CN115001057A - Composite micro-energy system and energy control method, device and storage medium thereof - Google Patents

Abstract

An energy control method and device of a composite micro energy system, a storage medium and the composite micro energy system are provided, wherein the composite micro energy system comprises a micro energy collection module, an energy storage module and a power supply module, and the method comprises the following steps: identifying characteristic information of each energy collected by the micro-energy collection module; inputting the characteristic information of each energy and the electric power required by the load into a decision tree model to obtain a decision label of each energy, wherein the decision label is used for indicating the trend of each energy; controlling a trend of each energy according to the decision tag to transmit at least a portion of each energy to the power supply module and/or energy storage module. By adopting the scheme of the invention, the trend of various energies in the composite micro-energy system can be accurately determined, and the energy utilization rate is improved.

Description

Composite micro-energy system and energy control method, device and storage medium thereof

technical field

The invention relates to the technical field of energy, in particular to a composite micro-energy system and an energy control method, device and storage medium thereof.

Background technique

A composite micro-energy system is a system that can collect energy generated by multiple micro-energy sources, and store and/or release various energies (for example, to supply power to external loads). In the prior art, for the energy collected by the composite micro-energy system, a set of logic threshold value control strategies are artificially formulated to determine the energy trend according to theoretical analysis and engineering experience.

It can be understood that the external environment is variable and the energy collected by the composite micro-energy system is usually greatly affected by the external environment. For example, seasonal changes and day and night changes may cause the solar energy collected by the system to change all the time. In addition, the needs of users who use composite micro-energy sources are also variable. For example, the energy required by external loads in different situations usually varies greatly. Since the logic threshold value control strategy in the prior art is artificially formulated, the cost of changing the logic threshold value control strategy is relatively high, so this artificially formulated logic threshold value control strategy is rarely changed once it is determined, so the existing The method in the technology cannot make the energy trend in the composite micro-energy system adapt to different situations such as the changing external environment and actual demand.

Therefore, there is an urgent need for an energy control method for a composite micro-energy system, which can more accurately determine the direction of various energies in the composite micro-energy system under different conditions and improve energy utilization.

SUMMARY OF THE INVENTION

The technical problem solved by the present invention is to provide an energy control method for a composite micro-energy system, which can more accurately determine the trends of various energies in the composite micro-energy system under different conditions and improve energy utilization.

In order to solve the above technical problems, an embodiment of the present invention provides an energy control method for a composite micro-energy system, the composite micro-energy system includes a micro-energy collection module, an energy storage module and a power supply module, and the micro-energy collection module is used to collect At least one kind of energy, the energy storage module is used to store the electric energy after the energy conversion, the power supply module is used to supply power to the load, and the method includes: identifying the characteristics of each energy collected by the micro energy collection module information; input the characteristic information of each energy and the electric power required by the load into the decision tree model to obtain the decision label of each energy, the decision label is used to indicate the direction of each energy; control according to the decision label the direction of each energy, so as to transmit at least a part of each energy to the power supply module and/or the energy storage module; wherein, the decision tree model is generated by training using a plurality of first energy samples as training data.

Optionally, the feature information includes one or more of the following: the type of energy and the electric power that the energy can be converted into.

Optionally, the method for obtaining a decision label for each energy further includes: inputting the state of charge of the energy storage module together with the characteristic information of each energy and the electrical power required by the load into the decision tree model, to get the decision label for each energy.

Optionally, the method for generating the decision tree model includes: acquiring a plurality of first energy samples, where each first energy sample includes electrical power required by the load, at least one sample energy and its characteristic information and decision label; A plurality of first energy samples are used as the training data, and the decision tree model is generated by training.

Optionally, using the plurality of first energy samples as the training data, before training to generate the decision tree model, the method further includes: determining a possible value range of the electrical power required by the load and each type of sample. The range of possible values of the energy characteristic information; according to the characteristic information of each sample energy in each first energy sample and the electric power required by the load corresponding to the first energy sample, the plurality of first energy samples are screened to obtain A first energy sample for generating the decision tree model is obtained.

Optionally, before inputting the characteristic information of each type of energy and the electrical power required by the load into the decision tree model, the method further includes: acquiring a plurality of second energy samples, each second energy sample including all the information of the load. The required power, at least one sample energy and its characteristic information, and a preset decision label, wherein the plurality of second energy samples and the plurality of first energy samples are independent of each other; The characteristic information of each sample energy and the electric power required by the load corresponding to the second energy sample are input into the decision tree model to obtain the model decision label of each sample energy in the second energy sample; The model decision label of each sample energy in the sample is compared with the preset decision label, and the number of second energy samples in which the model decision label of each sample energy is calculated to be consistent with the preset decision label accounts for all the first energy samples. The ratio of the number of energy samples, if the ratio is smaller than the first preset threshold, the decision tree model is pruned to update the decision tree model.

Optionally, the energy storage module includes a plurality of energy storage elements, the energy storage module is further configured to supply power to the load, and the method further includes: acquiring the state of charge of the plurality of energy storage elements; The electric power to be provided by the energy storage module and the state of charge of the plurality of energy storage elements are obtained by adopting a fuzzy control algorithm to obtain the power supply ratio of the plurality of energy storage elements, wherein the electric power to be provided by the energy storage module is determined by the load. The difference between the required electrical power and the energy convertible electrical power of the micro-energy collection module; the output electrical power of each energy storage element is determined according to the power supply ratio and the electrical power required by the load.

Optionally, the plurality of energy storage elements include super capacitors.

Optionally, according to the electric power to be provided by the energy storage module and the state of charge of the plurality of energy storage elements, using a fuzzy control algorithm to obtain the power supply ratio of the plurality of energy storage elements includes: The electric power of the energy storage module determines the power range to which the electric power to be provided by the energy storage module belongs, and determines the charging range to which the state of charge of each energy storage element belongs according to the state of charge of each energy storage element; Query the fuzzy control rule table for the charge range of each energy storage element to determine the power supply ratio range of each energy storage element; defuzzify the power supply ratio range of each energy storage element to obtain the power supply ratio of each energy storage element ; wherein, the fuzzy control rule table is used to describe the mapping relationship between the power range and the charge range and the power supply ratio range.

Optionally, the micro-energy collection module includes a plurality of micro-energy collectors, and the method further includes: acquiring first feedback information, where the first feedback information is used to indicate whether the micro-energy collectors are replaced; The first feedback information judges whether the micro-energy collector is replaced, and if so, obtains the energy convertible electric power of each replaced micro-energy collector; The converted electric power is compared with the second preset threshold; if the energy convertible electric power of any of the replaced micro-energy collectors does not exceed the second preset threshold, the decision tree model is retrained to generate the decision tree model.

Optionally, the energy storage module includes a plurality of energy storage elements, and the method further includes: acquiring second feedback information, where the second feedback information is used to indicate whether the energy storage element is replaced; The second feedback information judges whether the energy storage element is replaced, and if so, obtains the state of charge of each energy storage element after replacement; compares the upper limit of the state of charge of each energy storage element after replacement with the third preset Set a threshold for comparison; if the upper limit of the state of charge of any replaced energy storage element does not exceed the third preset threshold, retrain to generate the decision tree model.

In order to solve the above technical problems, an embodiment of the present invention also provides an energy control device for a composite micro-energy system, where the composite micro-energy system includes a micro-energy collection module, an energy storage module and a power supply module, and the micro-energy collection module is used for At least one type of energy is collected, the energy storage module is used to store the electrical energy converted from the energy, the power supply module is used to supply power to the load, and the device includes: an identification module for identifying the micro-energy collection module to collect electricity The characteristic information of each energy; the classification module is used to input the characteristic information of each energy and the electric power required by the load into the decision tree model, so as to obtain the decision label of each energy, and the decision label is used to indicate each energy. The direction of each energy; the transmission control module is used to control the direction of each energy according to the decision label, so as to transmit at least a part of each energy to the power supply module and/or the energy storage module; wherein, the decision tree The model is generated by using multiple first energy samples as training data.

An embodiment of the present invention further provides a storage medium on which a computer program is stored, and when the computer program is run by a processor, the steps of the energy control method for the composite micro-energy system described above are executed.

An embodiment of the present invention further provides a composite micro-energy system, the system includes: a micro-energy collection module, used to collect at least one type of energy; an energy storage module, used to store the converted electric energy; a power supply module, used for for supplying power to the load; the controller is used for executing the steps of the energy control method of the composite micro-energy system.

Compared with the prior art, the technical solutions of the embodiments of the present invention have the following beneficial effects:

In the solution of the embodiment of the present invention, since the decision tree model is generated by training using a plurality of first energy samples, the decision tree model generated by training can learn the characteristic information of the energy in the first energy samples, the power required by the load and the energy value of the load. Relationships and laws between decision labels. Input the identified feature information of each energy and the electrical power required by the load into the decision tree model generated by training, and the decision label for each energy can be determined according to the current feature information of each energy and the electrical power required by the load, and then based on The decision label controls the direction of each energy, so the direction of the energy can be controlled according to the characteristic information of each energy and the electric power required by the load. Through this method, the energy trend can be adapted to the characteristic information of the energy currently collected by the micro energy collection module and the electric power required by the load, so that the energy trend can be more accurately determined and the utilization rate of energy can be improved.

Further, in the solution of the embodiment of the present invention, before training to generate a decision tree model, a plurality of first energy samples are screened according to the determined range of possible values of the feature information and the range of possible values of the electric power required by the load, so as to obtain the available energy samples. For generating the first energy sample of the decision tree model, the accuracy of the first energy sample used as training data can be ensured, thereby making the decision tree model generated by training more accurate.

Further, in the solution of the embodiment of the present invention, before the user uses the decision tree model to control the direction of energy, a second energy sample independent of the first energy sample is used to test the decision tree model generated by training. When the ratio of the number of second energy samples whose model decision labels of each sample energy are consistent with the preset decision labels to the number of all second energy samples is less than the first preset threshold, the decision tree model is pruned to calibrate the Decision tree model, so that the decision tree model can adapt to the current actual demand, so as to more accurately determine the direction of energy.

Further, in the solution of the embodiment of the present invention, based on the fuzzy control algorithm, the power supply ratio of each energy storage element is determined according to the electric power to be provided by the energy storage module and the state of charge of each energy storage element, so that repeated use of individual energy storage elements can be avoided. The components supply power to the power supply module, so as to avoid problems such as premature aging and loss caused by repeated charging and discharging of the individual energy storage components as much as possible.

Description of drawings

FIG. 1 is a schematic structural diagram of a composite micro-energy system in an embodiment of the present invention.

FIG. 2 is a schematic flowchart of a first energy control method for a composite micro-energy system in an embodiment of the present invention.

FIG. 3 is a schematic structural diagram of a decision tree model in an embodiment of the present invention.

FIG. 4 is a schematic flowchart of a second energy control method for a composite micro-energy system in an embodiment of the present invention.

FIG. 5 is a schematic flowchart of a third energy control method for a composite micro-energy system in an embodiment of the present invention.

FIG. 6 is a schematic structural diagram of an energy control device of a composite micro-energy system in an embodiment of the present invention.

Detailed ways

As described in the background art, there is an urgent need for an energy control method for a composite micro-energy system, which can accurately determine the trends of various energies in the composite micro-energy system under different conditions and improve energy utilization.

As mentioned above, the inventors of the present invention have found through research that, in the prior art, the energy collected by the composite micro-energy system is usually determined by artificially formulating a set of logic threshold control strategies based on theoretical analysis and engineering experience. direction of energy. It can be understood that the external environment is variable and the energy collected by the composite micro-energy system is usually greatly affected by the external environment. For example, seasonal changes and day and night changes may cause the solar energy collected by the system to change all the time. In addition, the needs of users who use composite micro-energy sources are also variable. For example, the energy required by external loads in different situations usually varies greatly. Since the logic threshold value control strategy in the prior art is artificially formulated, the cost of changing the logic threshold value control strategy is relatively high, so this artificially formulated logic threshold value control strategy is rarely changed once it is determined, so the existing The method in the technology cannot make the energy trend in the composite micro-energy system adapt to different situations such as the changing external environment and actual demand.

In order to solve the above technical problem, the embodiment of the present invention provides a control method for a composite micro-energy system. In the solution of the embodiment of the present invention, since the decision tree model is generated by training using a plurality of first energy samples, the training generated The decision tree model can learn the relationship and law between the characteristic information of the energy in the first energy sample, the power required by the load and the decision label of the energy. Input the identified characteristic information of each energy and the electrical power required by the load into the decision tree model generated by training, and the decision label of each energy can be determined according to the current characteristic information of each energy and the electrical power required by the load, and then The direction of each energy is controlled according to the decision label, so that the direction of energy can be controlled according to the characteristic information of each energy and the electric power required by the load. Through this method, the energy trend can be adapted to the characteristic information of the energy currently collected by the micro energy collection module and the electric power required by the load, so that the energy trend can be accurately determined and the utilization rate of energy can be improved.

In order to make the above objects, features and beneficial effects of the present invention more clearly understood, specific embodiments of the present invention will be described in detail below with reference to the accompanying drawings.

Referring to FIG. 1, FIG. 1 is a schematic structural diagram of a composite micro-energy system in an embodiment of the present invention. The composite micro-energy system to which the embodiments of the present invention are applicable will be described in a non-limiting manner below with reference to FIG. 1 .

The composite micro-energy system shown in FIG. 1 may include: a micro-energy collection module 11 , an energy storage module 12 , a power supply module 13 and a controller 14 .

The micro-energy collection module 11 is used to collect at least one kind of energy, and the at least one kind of energy can be any kind of energy. The energy may be electrical energy, or any energy that can be converted into electrical energy, such as solar energy, electromagnetic energy, vibration energy, etc., but not limited thereto.

The micro-energy collection module 11 may include a plurality of micro-energy collectors, and the micro-energy collectors may be used to collect various forms of energy, and in addition, the various types of collected energy may be converted into electrical energy.

Specifically, the types of energy are different, and the microenergy harvesters are usually different. For example, micro energy harvesters may include solar energy harvesters (eg, solar panels), vibration energy harvesters, radio frequency energy harvesters, and the like. Among them, solar energy collectors are micro energy collectors for collecting solar energy, vibration energy collectors are micro energy collectors for collecting vibration energy, and radio frequency energy collectors are micro energy collectors for collecting electromagnetic energy. It is not limited to this.

The micro-energy collection module 11 can be coupled with the energy storage module 12 to transmit the collected energy (or converted electrical energy) to the energy storage module 12, or can be coupled with the power supply module 13 to transfer the collected energy (or converted electrical energy) to the energy storage module 12. The energy (or the converted electric energy) is transmitted to the power supply module 13, but is not limited thereto.

The energy storage module 12 is used to store the converted electric energy. Specifically, after the energy collected by the micro-energy collection module 11 is converted into electrical energy, it can be transmitted to the energy storage module 12 for storage. It should be noted that the energy stored by the energy storage module 12 is usually electrical energy.

The energy storage module 12 may include a plurality of energy storage elements, and the energy storage elements may refer to elements used to store electrical energy, such as, but not limited to, lithium batteries and/or super capacitors. The embodiments of the present invention do not impose any limitations on the quantity and types of energy storage elements.

The energy storage module 12 can be coupled with the power supply module 13 and can transmit the stored electrical energy to the power supply module 13 .

The power supply module 13 is used for coupling with an external load and supplying power to the external load. The electrical energy obtained by the power supply module 13 may come from the micro-energy collection module 11 , the energy storage module 12 , or both. In other words, the electrical energy obtained by the power supply module 13 may only come from the micro-energy collection module 11, only from the energy storage module 12, or both from the micro-energy collection module 11 and the energy storage module 12, but is not limited thereto.

The controller 14 can be used to control the direction of the energy collected by the micro-energy collection module in the composite micro-energy system.

Specifically, the controller 14 can be coupled with the micro-energy collection module 11 and used to control the direction of each energy collected by the micro-energy collection module 11 , and more specifically, the controller 14 can be used to control the micro-energy collection module 11 Whether the collected energy is transmitted to the energy storage module 12, to the power supply module 13, or to both at the same time.

Further, the controller 14 may also be coupled to the energy storage module 12 and used to control whether the electrical energy stored in the energy storage module 12 needs to be transmitted to the power supply module 13 . More specifically, the controller 14 may also be coupled with each energy storage element in the energy storage module 12 and used to control the electrical energy transmitted by each energy storage element to the power supply module 13 .

Further, the controller 14 can also be coupled with the power supply module 13 to obtain the current power supply demand, where the power supply demand refers to the electric power that the composite micro-energy system currently needs to provide to the outside, for example, the electric power required by the load.

The composite micro-energy system may further include a monitoring module 15, and the monitoring module 15 may be coupled with the micro-energy collection module 11 to monitor whether the micro-energy collection module 11 is abnormal. More specifically, the monitoring module 15 can be used to monitor whether each micro-energy collector is abnormal, for example, whether various parameters of the micro-energy collector (such as parameters such as the peak power of output energy) are abnormal, and whether the micro-energy collector itself exists. Cracks, damages, occlusions, etc.

Further, the monitoring module 15 may also be coupled with the energy storage module 12 to monitor whether the energy storage module 12 is abnormal. More specifically, the monitoring module 15 can be used to monitor whether each energy storage element in the energy storage module 12 is abnormal, for example, whether various parameters of the energy storage element (such as parameters such as the upper limit of the stored energy) are abnormal, and the energy storage element itself is aging. Wait.

Referring to FIG. 2 , FIG. 2 is a schematic flowchart of an energy control method for a composite micro-energy system in an embodiment of the present invention. The energy control method can be performed by a controller, and the controller can be any appropriate terminal with data receiving and processing capabilities, such as a computer, a sensor analyzer, etc., but not limited thereto. The controller can be arranged inside the composite micro-energy system, and the controller is coupled to each device in the composite micro-energy system; or, the controller can be arranged outside the composite micro-energy system and remotely connected to the composite micro-energy system. The respective devices are coupled, but not limited thereto. The composite micro-energy system can be a system that can collect the energy generated by a variety of micro-energy sources, and store and/or release various energies. The composite micro-energy system can include a micro-energy collection module, an energy storage module and a power supply module. , but not limited to this. In a non-limiting embodiment of the present invention, the composite micro-energy system is the composite micro-energy system shown in FIG. 1 .

The energy control method of the composite micro-energy system shown in FIG. 2 may include the following steps:

Step S201: Identify the characteristic information of each type of energy collected by the micro-energy collection module;

Step S202: Input the characteristic information of each energy and the electric power required by the load into the decision tree model to obtain the decision label of each energy, the decision label is used to indicate the trend of each energy; wherein, the decision The tree model is generated by using multiple first energy samples as training data.

Step S203: Control the direction of each energy according to the decision label, so as to transmit at least a part of each energy to the power supply module and/or the energy storage module.

In the specific implementation of step S201, characteristic information of each energy collected by the micro-energy collection module may be acquired and identified according to a preset time interval, and the preset time interval may be pre-configured. The feature information refers to features used to describe one or more aspects of the energy. For example, the characteristic information may include the type of energy and the electric power to which the energy can be converted, but is not limited thereto.

Specifically, the micro-energy collection module may include a plurality of micro-energy collectors. Since the types of energy are different, the micro-energy collectors are also different, so the type of energy can be determined according to the micro-energy collector from which each energy comes. More specifically, each micro-energy harvester may have an energy identifier, micro-energy harvesters that collect the same type of energy may have the same energy identifier, and the type of energy may be determined by identifying the energy identifier of each micro-energy harvester.

Further, the convertible electric power of each energy in the micro-energy collection module can also be obtained. The convertible electrical power of the energy refers to the efficiency with which the energy can be converted into electrical energy. Specifically, the convertible electric power of each energy in the micro-energy collection module can be obtained through components such as sensors installed in the micro-energy collection module. More specifically, the energy-convertible electrical power in each micro-energy harvester can also be obtained.

In the specific implementation of step S202, the electric power required by the external load can be obtained according to a preset time interval, and the characteristic information of each energy and the electric power required by the load are input into the decision tree model to obtain the decision label of each energy.

Wherein, the decision label can be used to indicate the direction of energy, for example, to the power supply module and/or to the energy storage module. In a non-limiting embodiment of the present invention, the energy storage element in the energy storage module includes a lithium battery and a super capacitor, and the direction of energy may include one or more of the following: transmission to the power supply module, transmission to the lithium battery, and transmission to supercapacitors.

It should be noted that the relationship between the characteristic information of the energy, the electric power required by the load and the decision label is related to time, and the relationship may change with time. Specifically, the characteristic information of the energy and the electric power required by the load may belong to the same time interval, and the obtained decision label is also the trend of the energy in the time interval. That is, every time interval, according to the current characteristic information of each energy and the electric power required by the load, the current decision label of each energy is obtained.

In a non-limiting embodiment of the present invention, the state of charge of the energy storage module can be input into the decision tree model together with the characteristic information of each energy and the electric power required by the load to obtain the decision label of each energy. In other words, the decision label for each energy is determined according to the state of charge of the energy storage module, the characteristic information of the energy, and the electrical power required by the load. Among them, the state of charge of the energy storage module and the characteristic information of the energy, the electric power required by the load and the decision label also have a time-related correlation. Compared with the scheme of determining the decision label of each energy according to the characteristic information of each energy and the electric power required by the load, the state of charge of the energy storage module, the characteristic information of each energy, and the electric power required by the load are input to the decision-making The tree model is used to obtain the category label decision label scheme of each energy, which can make the energy trend more in line with the actual needs of the composite micro-energy system and improve the utilization rate of energy.

More specifically, the state of charge of each energy storage element in the energy storage module can be input into the decision tree model together with the characteristic information of the energy and the electric power required by the load. That is, the decision label of the energy may be determined by the characteristic information of the energy having a time-related relationship with it, the electric power required by the load, and the state of charge of each energy storage element.

Further, the decision tree model is generated by training using a plurality of first energy samples as training data. Wherein, each first energy sample may include electrical power required by the load, at least one sample energy, and characteristic information and decision labels of each sample energy. The sample energy can be selected from the energy collected by the micro-energy collection module, for example, solar energy, vibration energy, electromagnetic energy, etc., but not limited thereto. Among them, each kind of sample energy has characteristic information and decision label, and the characteristic information and decision label of different sample energy can be different.

It should be noted that the electrical power required by the load included in each first energy sample has a time-related correlation with the first energy sample. More specifically, the electrical power required by the load and each sample energy in the first energy sample have a time-dependent relationship.

In one non-limiting embodiment of the present invention, each first energy sample may also include a state of charge of the energy storage module. The state of charge of the energy storage module and the first energy sample may also have a time-dependent correlation. More specifically, the state of charge of the energy storage module may include the state of charge of a plurality of energy storage elements. It should be noted that the specific content of the first energy sample depends on the input data of the decision tree model.

Further, for each combination of sample energy, the multiple first energy samples need to include all possible trends of each energy in this combination. The combination mode refers to the type of sample energy in each first energy sample. Taking the combination of solar energy and vibrational energy as an example, in the first energy sample where the sample energy is solar energy and vibrational energy, the decision labels that need to include solar energy are the first energy samples of all optional decision labels, and the first energy samples of vibrational energy. The decision labels are the first energy samples of all optional decision labels, respectively.

In a non-limiting embodiment of the present invention, considering that some of the first energy samples may be obtained when the composite micro-energy system is abnormal, in order to ensure the accuracy of the decision tree model, it is necessary to energy samples for screening.

Specifically, the range of possible values of the characteristic information of each sample energy and the range of possible values of the electrical power required by the load may be determined first. The aforementioned range of possible values may be received from the outside, or may be pre-stored in the controller. Wherein, the possible value range of the characteristic information of each sample energy may be the possible value range of the energy identifier, or the possible value range of the energy convertible electric power of each energy, but is not limited thereto.

Further, it is judged whether the characteristic information of each first energy sample satisfies the acceptable value range of the characteristic information, and whether the electric power required by the load corresponding to each first energy sample satisfies the acceptable value range of the electric power required by the load, if both If both are satisfied, it can be judged that the first energy sample can be used for training a decision tree model.

In a non-limiting embodiment of the present invention, the range of the state of charge of the energy storage module can also be determined, and the plurality of first energy samples are screened according to the state of charge of the energy storage module corresponding to the first energy sample to obtain A first energy sample for generating the decision tree model.

Therefore, by setting the possible value range of the feature information and the possible value range of the electric power required by the load, a plurality of first energy samples can be screened, the accuracy of the training data is improved, and the decision tree model is more accurate.

Further, the method of using a plurality of first energy samples to train and generate a decision tree model can be any existing appropriate algorithm, such as a classification and regression tree (Classification and Regression Tree, CART) algorithm, that is, can be based on Gini ( Gini) exponential minimization principle to recursively build decision tree models.

Referring to FIG. 3, FIG. 3 is a schematic diagram of a decision tree model in an embodiment of the present invention. The decision tree model shown in FIG. 3 includes nodes and directed edges, wherein the nodes include internal nodes 31 and leaf nodes 32 . The internal node 31 represents a variety of preset attributes, and the preset attributes can be determined according to the input of the decision tree model. For example: the type of energy, the convertible electric power of the energy and the electric power required by the load, etc. Leaf nodes 32 may represent multiple decision labels.

Wherein, each internal node 31 corresponds to an attribute test, and each internal node 31 contains a plurality of first energy samples which are divided into sub-nodes of the internal node 31 according to the result of the attribute test. Starting from the uppermost internal node in the internal nodes 31, the multiple first energy samples contained in each internal node 31 are recursively classified until the preset conditional position for stopping the recursion is reached, so as to obtain multiple leaves Node 32. The preset condition for stopping the recursion may be that the number of multiple first energy samples in each internal node 31 is less than the preset number, or the Gini index of multiple first energy samples included in each internal node 31 is less than Preset indices, etc., but not limited thereto.

Specifically, for the sample set D included in each internal node 31, the sample set D includes a plurality of first energy samples, the sample set D has K sample subsets, and the Gini index of the sample set D is calculated by the following formula:

是样本集D中的第一能量样本划入样本子集D_k的概率。where Gini(D) is the Gini index of the sample set D, K is the number of child nodes of the internal node 31,

is the probability that the first energy sample in the sample set D is classified into the sample subset _Dk .

Further, there are multiple possible values of attribute A corresponding to each internal node 31, and for each possible value a, the test result of attribute A=a according to each first energy sample is “Yes” or "No", divide the sample set D into sample subset D ₁ and sample subset D ₂ , and use the following formula to calculate the Gini index of sample set D when A=a:

为样本集D中的第一能量样本划入样本子集D₁的概率，

为样本集D中的第一能量样本划入样本子集D₂的概率。基尼指数Gini(D,a)可以表示根据第一能量样本的属性A取值为a来分割样本集D时的不确定性，基尼指数Gini(D,a)越大，这种分割方式的不确定性就越大。Among them, Gini(D,a) is the Gini index when the value of the attribute A of the internal node 31 is a, Gini(D ₁ ) is the Gini index of the sample subset D ₁ , Gini(D ₂ ) is the sample subset D Gini index of ₂ ,

is the probability that the first energy sample in sample set D is classified into sample subset D ₁ ,

The probability that the first energy sample in sample set D will be classified into sample subset D ₂ . The Gini index Gini(D,a) can represent the uncertainty when dividing the sample set D according to the value of the attribute A of the first energy sample as a. The larger the Gini index Gini(D,a) is, the greater the difference in this way of dividing. The greater the certainty.

For different values a of attribute A, the corresponding Gini indices Gini(D, a) are calculated respectively, and the value a with the smallest Gini index is selected, and the first energy sample of the internal node 31 is divided into A sample set of two child nodes.

Further, the above steps are recursively invoked on the two child nodes until the preset conditions for stopping the recursion are met, thereby generating a decision tree model.

Further, in the solution of the embodiment of the present invention, after the decision tree model is generated by training, the obtained decision tree model can also be tested, and the test method can be any appropriate method that can be used to test the accuracy of the generated decision tree model, such as , the S-fold cross-validation method can be used for testing, but it is not limited to this.

Specifically, before training the decision tree model, a plurality of first energy samples can be divided into a training sample set and a test sample set, and after the decision tree model is generated by training the first energy samples in the training sample set, each The electric power required by the load of the first energy sample and the characteristic information of each sample energy are input into the decision tree model to obtain the model decision label of each sample energy in the first energy sample. Compare the model decision label of each sample energy in each first energy sample with the decision label of the sample energy, and calculate the number of first energy samples whose model decision label of each sample energy is consistent with the decision label accounts for all the first energy samples. A ratio of the number of energy samples. If the ratio is less than the preset ratio, it is judged that the decision tree model is inaccurate, and the decision tree model needs to be processed to improve the accuracy of the decision tree model. For example, the decision tree model can be pruned by Pruning can make the decision tree model more compact and prevent overfitting.

In a non-limiting embodiment of the present invention, before inputting the characteristic information of the energy and the electrical power required by the load into the decision tree model, the decision tree model may also be calibrated. Specifically, the decision tree model is calibrated using a second energy sample that is independent of the first energy sample. Among them, "independent" means that there is no correlation between the first energy sample and the second energy sample. For example, the first energy sample is the sample selected when training the decision tree model, and the second energy sample is the sample before the actual use of the decision tree model. The sample selected during calibration. More specifically, in different usage scenarios, the electrical power required by the load varies greatly, and the training data of the decision tree model cannot cover the electrical power required by the load in all scenarios. Before the trend is controlled, a second energy sample can be selected according to the actual usage scenario to calibrate the decision tree model.

Specifically, a plurality of second energy samples may be obtained, and each second energy sample includes electrical power required by the load and at least one sample energy, its characteristic information, and a preset decision label. For more description about the second energy sample, reference may be made to the above related description about the first energy sample, which will not be repeated here.

Further, the characteristic information of each kind of sample energy in each second energy sample and the electric power required by the load corresponding to the second energy sample can be input into the decision tree model to obtain each kind of sample in the second energy sample. Model decision label for energy. It should be noted that, if the input of the decision tree model also includes the state of charge of the energy storage module, the state of charge of the energy storage module corresponding to each second energy sample needs to be input into the decision tree model together.

Further, compare the model decision label of each kind of sample energy in each second energy sample with the preset decision label, and calculate the consistency between the model decision label of each kind of sample energy and the preset decision label. The ratio of the number of second energy samples to the number of all second energy samples, if the ratio is smaller than the first preset threshold, the decision tree model is pruned to update the decision tree model. Wherein, the first preset threshold may be pre-configured. The method of the pruning treatment may be any of existing appropriate methods.

From the above, in the solution of the embodiment of the present invention, before the user uses the decision tree model to control the direction of energy, the decision tree model is pruned to calibrate the decision tree model, so that the decision tree model can be adapted to the current actual situation. demand to more accurately determine the direction of energy.

Continuing to refer to FIG. 2, in the specific implementation of step S203, after obtaining the decision label of each energy, the direction of each energy can be controlled according to the decision label of the energy, so as to transmit at least a part of each energy to the power supply module and/or or energy storage modules.

Specifically, a controllable path may be provided between the micro-energy collection module, the power supply module, and the energy storage module. The corresponding control signal can be generated according to the decision label of each energy. The control signal can be used to set the path between the micro-energy collection module and the power supply module to be turned on or off, or used to set the micro-energy collection module and the storage module. The paths between the modules can be turned on or off. More specifically, the control signal can be used to set the path between each micro energy collector and the power supply module to be on or off, and can also be used to set the path between each energy storage element and the power supply module to be on or off.

Therefore, in the solution of the embodiment of the present invention, the identified characteristic information of each energy and the electric power required by the load are input into the trained decision tree model, and the current characteristic information of each energy and the electric power required by the load can be input according to the current To determine the decision label of each energy, and then control the direction of each energy according to the decision label, so that the energy direction can be controlled according to the current characteristic information of each energy and the electrical power required by the load. Such a control process can be executed once every preset time. Through this method, the energy trend can be adapted to the characteristic information of the energy currently collected by the micro-energy collection module and the electric power required by the load, so that the energy can be accurately Determine the direction of energy, all kinds of energy can be fully utilized, and the utilization rate of energy is improved.

Referring to FIG. 4 , FIG. 4 shows a second energy control method for a composite micro-energy system in an embodiment of the present invention. The method is used to control the power supply ratio of each energy storage element in the energy storage module to the power supply module. It should be noted that the energy storage module can also be used to transmit electrical energy to the power supply module.

Preferably, when the power supply module supplies power to the load, the electrical energy collected in the micro-energy collection module can be firstly transmitted to the power supply module, and when the electrical power that can be converted from the energy in the micro-energy collection module is lower than the electric power threshold, the micro-energy collection module no longer Power is transmitted to the power supply module, and the energy storage module can continue to transmit power to the power supply module at this time. This method of supplying power from the micro-energy collection module first and then the energy storage module can avoid problems such as rapid aging caused by frequent charging and discharging of the energy storage elements in the energy storage module.

The energy control method of the composite micro-energy system shown in FIG. 4 may include the following steps:

Step S401: acquiring the state of charge of a plurality of energy storage elements;

Step S402: According to the electric power to be provided by the energy storage module and the state of charge of the plurality of energy storage elements, a fuzzy control algorithm is used to obtain the power supply ratio of the plurality of energy storage elements, wherein the energy storage module needs to provide the power supply ratio. The electric power is the difference between the electric power required by the load and the energy-convertible electric power of the micro-energy collection module;

Step S403: Determine the output electric power of each energy storage element according to the power supply ratio and the electric power to be provided by the energy storage module.

In the specific implementation of step S401, the state of charge (State of Charge, SoC) of each energy storage element in the energy storage module may be acquired. Preferably, the energy storage element may include a supercapacitor, and the supercapacitor has the advantage of fast charging and discharging. It is understandable that the long-term repeated high-current charging and discharging of the lithium battery will lead to the rapid aging of the lithium battery. Compared with the scheme that the energy storage element only includes the lithium battery, the use of the super capacitor as the energy storage element can make the super capacitor and the lithium battery. can cooperate effectively.

In the specific implementation of step S402, the electric power to be provided by the energy storage module and the state of charge of each energy storage element are respectively fuzzified. Specifically, the power range to which the electric power to be provided by the energy storage module belongs can be determined according to the electric power to be provided by the energy storage module, and the state of charge of each energy storage element is determined according to the state of charge of the energy storage element the charge range it belongs to.

Among them, various existing appropriate algorithms can be used to fuzzify the electric power to be provided by the energy storage module and the state of charge of each energy storage element. Preferably, the membership value method can be used for fuzzification.

Specifically, the respective membership functions of the electric power to be provided by the energy storage module and the state of charge of each energy storage element may be determined first, and the interval of the membership functions may be uniform or non-uniform. The membership function may be a double sigmoid membership function, a triangular membership function, a Gaussian membership function, a sigmoid membership function, a trapezoidal membership function, etc., but is not limited thereto. Then, the power range is determined according to the electric power to be provided by the energy storage module and its corresponding membership function, and the charge range corresponding to the energy storage element is determined according to the state of charge of each energy storage element and its membership function.

It should be noted that the electric power to be provided by the energy storage module is the difference between the electric power required by the load and the energy-convertible electric power of the micro-energy collection module. More specifically, when the electrical power that can be converted into energy in the micro-energy collection module is lower than the electric power threshold, the micro-energy collection module no longer transmits electrical energy to the power supply module. In order to prolong the service life of the devices in the micro-energy collection module, the micro-energy The energy convertible electric power of the collection module is subtracted from a preset fixed value, and then used to calculate the electric power to be provided by the energy storage module, and the preset fixed value may be the electric power threshold.

Further, query the fuzzy control rule table according to the power range and the charge range of each energy storage element to determine the power supply ratio range of each energy storage element. The fuzzy control rule table may be pre-configured to describe the mapping relationship between the power range and the charge range and the power supply ratio range. The fuzzy control rule table may include a set of fuzzy conditional statements with an if-then structure, so as to query the corresponding power supply ratio range from the input power range and charge range.

Further, the power supply ratio range of each energy storage element is defuzzified to obtain the power supply ratio of each energy storage element. In other words, the power supply ratio of each energy storage element is determined according to the power supply ratio range of the energy storage element. The method of defuzzification can be any existing appropriate algorithm, such as maximum membership degree method, weighted average method, center of gravity method, etc., but is not limited thereto. Preferably, an appropriate method can be selected according to the requirements of the composite micro-energy system or the actual situation of operation, so as to convert the ambiguous power supply ratio range into an accurate power supply ratio. More specifically, the defuzzification method can be selected according to the performance requirements of the composite micro-energy system. For example, when the composite micro-energy system has high requirements for the calculation speed of determining the power supply ratio, the maximum membership degree method and the weighted average method can be selected. ; When the composite micro-energy system requires high calculation accuracy for determining the power supply ratio, the center of gravity method can be selected, but it is not limited to this. In a non-limiting embodiment of the present invention, the gravity center method is used to defuzzify the power supply ratio range of each energy storage element, so as to obtain the power supply ratio of each energy storage element.

In the specific implementation of step S403, the output electric power of each energy storage element can be calculated and determined according to the power supply ratio of each energy storage element and the electric power to be provided by the negative energy storage module, and the output electric power of each energy storage element can be determined according to the output of each energy storage element. The electrical power supplies power to the power supply module. From the above, in the solution of the embodiment of the present invention, the purpose of extending the service life of the system can be achieved while satisfying the power supply requirement of the load.

Referring to FIG. 5 , FIG. 5 shows a third energy control method for a composite micro-energy system in an embodiment of the present invention. Compared with the energy control method of the composite micro-energy system shown in FIG. 2 , the energy control method of the composite micro-energy system shown in FIG. 5 may further include the following steps:

Step S204: obtaining first feedback information, where the first feedback information is used to indicate whether the micro energy collector is replaced;

Step S205: Determine whether the micro-energy collector is replaced according to the first feedback information, and if so, obtain the energy convertible electric power of each replaced micro-energy collector;

Step S206: Compare the energy convertible electrical power of each replaced micro-energy device collector with the second preset threshold, if the energy convertible electrical power of any replaced micro-energy device collector does not exceed the first Two preset thresholds, retrain to generate the decision tree model.

In the specific implementation of step S204, the energy-convertible electric power in each micro-energy collector can be obtained, and the energy-convertible electric power in each micro-energy collector can be compared with a preset second preset threshold, when any one When the energy convertible electric power in the micro-energy collector does not exceed the second preset threshold, it can be determined that the micro-energy collector is abnormal. More specifically, when the energy-convertible electric power in any one of the micro-energy collectors does not continue to exceed the second preset threshold for longer than the first preset time, it can be determined that the micro-energy collector is abnormal. The second preset threshold and the first preset time may be preconfigured.

Further, when it is determined that the micro-energy collection device is abnormal, first alarm information may be issued, and the first alarm information may indicate the abnormal micro-energy collection device, so that the user can replace it.

Further, the first feedback information can be received from the outside, and the first feedback information can be used to indicate whether the user replaces the micro energy collector.

In the specific implementation of step S205, it can be determined whether the micro energy collector is replaced according to the received first feedback information. In addition, if the first feedback information is not obtained after the second preset time, it can be determined that the micro-energy collector has not been replaced.

Further, after it is determined that the micro-energy collector is replaced, the energy convertible electric power of the replaced micro-energy collector can be obtained. Specifically, what is obtained may be the upper limit of the electrical power that can be converted into energy within the first preset time after the micro-energy collector is replaced.

In the specific implementation of step S206, the energy convertible electrical power of the replaced micro-energy collector can be compared with the second preset threshold, if the energy convertible electrical power of any replaced micro-energy collector does not exceed the With the second preset threshold, the decision tree model can be retrained to generate the decision tree model, so that the decision tree model can adapt to the composite micro-energy system after the replacement of the micro-energy collector, and realize the self-correction of the decision tree model.

Therefore, in the solution of the embodiment of the present invention, it is not necessary to retrain the generated decision tree model after judging that the micro-energy collector is replaced, but firstly, the energy convertible electric power and the second preset threshold are compared again to determine the replacement. Whether there is still an abnormal situation after replacing the micro energy collector, if the abnormal situation disappears after replacing the micro energy collector, there is no need to retrain the generated decision tree model. The scheme in is more efficient.

In addition, compared with not regenerating the decision tree model after replacing the micro-energy collector, in the solution of the embodiment of the present invention, the replaced micro-energy collector is monitored again, so as to ensure timely discovery of whether the replaced micro-energy collector exists. abnormal, so as to ensure the normal operation of the composite micro-energy system.

In a non-limiting embodiment of the present invention, if it is determined according to the first feedback information that the micro energy collector has not been replaced, the decision tree model can also be retrained to generate a decision tree model, so that the decision tree model can adapt to the abnormal situation of the micro energy collector , to realize the self-correction of the decision tree model.

Specifically, a plurality of third energy samples may be obtained, and the third energy samples may be used to retrain the generated decision tree model. Wherein, each third energy sample may include the power required by the load, at least one sample energy and its characteristic information and decision label. Wherein, the characteristic information of the sample energy in each third energy sample meets the characteristic requirements of the energy collected by the micro energy collector when the micro energy collector is abnormal. For example, the energy-convertible electrical power of the sample energy in each third energy sample does not exceed the second preset threshold. For more specific content about using the third energy sample to retrain and generate the decision tree model, please refer to the above related description about using the first energy sample to train and generate the decision tree model, which will not be repeated here.

In another non-limiting embodiment of the present invention, considering that the model, type, etc. of the micro-energy collector may change before and after the replacement, so that the original decision tree model is not suitable for the replaced micro-energy collector, it is also possible to calculate The difference between the energy convertible electric power of each micro-energy collector after replacement and the standard value of the energy-convertible electric power of the micro-energy collector, and comparing the difference with a preset first difference threshold, If the difference is greater than the first difference threshold, it can indicate that there is a big difference between the micro-energy collector after replacement and the micro-energy collector before replacement, and the decision tree model can be generated by retraining.

Wherein, the standard value of the energy convertible electric power may be determined according to the micro energy collector. The standard value of the energy convertible electric power of different micro-energy harvesters can be different. More specifically, the standard value of the energy convertible electric power of different types of micro-energy harvesters may be different. For the replaced micro-energy collector, the standard value of its energy convertible electric power may be calculated according to the energy convertible electric power of the micro-energy collector within the third preset time before replacement. For example, the middle value of the energy-convertible electric power within the third preset time may be taken as the standard value of the energy-convertible electric power, but it is not limited thereto.

In another non-limiting embodiment of the present invention, the state of charge of each energy storage element can also be acquired, and the state of charge of each energy storage element is compared with a preset third preset threshold, when When the state of charge of any one of the energy storage elements does not exceed the third preset threshold, it can be determined that the energy storage element is abnormal. More specifically, when the state of charge of any one of the energy storage elements does not continue to exceed the third preset threshold for longer than the first preset time period, it may be determined that the energy storage element is abnormal. The third preset threshold may be preconfigured.

Further, when it is determined that there is an abnormality in the energy storage element, second alarm information may be issued, and the second alarm information may indicate the abnormal energy storage element, so that the user can replace it.

Further, the second feedback information can be received from the outside, and the second feedback information can be used to indicate whether the user replaces the energy storage element.

Further, whether the energy storage element is replaced can be determined according to the received second feedback information. In addition, if the second feedback information is not obtained after the second preset time, it can be determined that the energy storage element has not been replaced.

Further, after it is determined that the energy storage element is replaced, the state of charge of the replaced energy storage element can be obtained. Specifically, the upper limit of the state of charge within the first preset time after the energy storage element is replaced can be obtained. More specifically, the upper limit of the state of charge of the energy storage element within the first preset time can be obtained.

Further, the state of charge of each replaced energy storage element can be compared with the third preset threshold, and if the state of charge of any replaced energy storage element does not exceed the third preset threshold , the decision tree model can be retrained to generate the decision tree model.

In a non-limiting embodiment of the present invention, considering that the models, types, etc. of the energy storage elements may change before and after replacement, the original decision tree model is not suitable for the replaced energy storage elements, and the replacement energy storage elements can also be calculated. The difference between the state of charge of each energy storage element and the standard value of the state of charge of the energy storage element, and the difference is compared with a preset second difference threshold, if the difference is greater than the second The difference threshold value can indicate that there is a big difference between the energy storage element after replacement and the energy storage element before replacement. At this time, the decision tree model can be retrained.

Wherein, the standard value of the state of charge may be determined according to the energy storage element. The standard value of the state of charge of different energy storage elements can be different. More specifically, the standard value of the state of charge of different types of energy storage elements may be different. For the energy storage element to be replaced, the standard value of the state of charge of the energy storage element may be calculated according to the state of charge of the energy storage element within the third preset time before the replacement, for example, the third preset time may be taken. The intermediate value of the state of charge within the SOC is used as the standard value of the state of charge, but is not limited to this.

For more specific content of judging whether to retrain the generated decision tree model according to the second feedback information, reference may be made to the above related description about FIG. 5 , which will not be repeated here.

Referring to FIG. 6, FIG. 6 is an energy control device of a composite micro-energy system in an embodiment of the present invention, and the device may include:

The identification module 61 is used to identify the characteristic information of each kind of energy collected by the micro-energy collection module;

The classification module 62 is used for inputting the characteristic information of each energy and the electric power required by the load into the decision tree model to obtain the decision label of each energy, and the decision label is used to indicate the trend of each energy;

a transmission control module 63, configured to control the direction of each energy according to the decision label, so as to transmit at least a part of each energy to the power supply module and/or the energy storage module;

Wherein, the decision tree model is generated by using a plurality of first energy samples as training data for training.

For the principle, working mode and beneficial effects of the energy control device of the composite micro-energy system in the embodiment of the present invention, please refer to the relevant description of the energy control method of the composite micro-energy system above, which will not be repeated here.

Referring to FIG. 1 , an embodiment of the present invention further provides a composite micro-energy system. The system may include: a micro-energy collection module 11 for collecting at least one type of energy; an energy storage module 12 for storing the energy after conversion The power supply module 13 is used to supply power to the load; the controller 14 is used to execute the steps of the energy control method of the composite micro-energy system.

Wherein, the controller can be coupled with a memory storing a computer program, the controller can read the computer program in the memory, and execute the steps of the energy control method of the composite micro-energy system by running the computer program. It should be noted that the controller may be a processor independent of the memory, or may be a terminal integrated with the memory and the processor, but is not limited thereto.

For the principle, structure, working mode, and beneficial effects of the composite micro-energy system in the embodiments of the present invention, please refer to the foregoing related description of the energy control method of the composite micro-energy system, which will not be repeated here.

The embodiment of the present invention also discloses a storage medium, which is a computer-readable storage medium, and stores a computer program thereon, and the computer program can execute the steps of the above method when running. The storage medium may include ROM, RAM, magnetic or optical disks, and the like. The storage medium may also include a non-volatile memory (non-volatile) or a non-transitory (non-transitory) memory and the like.

The processor may be a central processing unit (CPU for short), and the processor may also be other general-purpose processors, digital signal processors (DSP), application specific integrated circuits (application specific integrated circuits). integrated circuit, referred to as ASIC), ready-made programmable gate array (field programmable gate array, referred to as FPGA) or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components and the like. A general purpose processor may be a microprocessor or the processor may be any conventional processor or the like.

It should be noted that the controller may further include memory, which may be volatile memory or non-volatile memory, or may include both volatile and non-volatile memory. Among them, the non-volatile memory may be a read-only memory (Read-only Memory, referred to as ROM), a programmable read-only memory (Programmable ROM, referred to as PROM), erasable programmable read-only memory (Erasable PROM, referred to as EPROM) , Electrically Erasable Programmable Read-Only Memory (Electrically EPROM, EEPROM for short) or flash memory. The volatile memory may be random access memory (RAM for short), which is used as an external cache. By way of example and not limitation, many forms of random access memory (RAM) are available, such as static random access memory (SRAM), dynamic random access memory (DRAM), synchronous Dynamic random access memory (Synchronous DRAM, referred to as SDRAM), double data rate synchronous dynamic random access memory (Double datarate SDRAM, referred to as DDR SDRAM), enhanced synchronous dynamic random access memory (Enhanced SDRAM, referred to as ESDRAM), synchronous A dynamic random access memory (Synchlink DRAM, referred to as SLDRAM) and a direct memory bus random access memory (Direct rambus RAM, referred to as DR RAM) are connected.

The above embodiments may be implemented in whole or in part by software, hardware, firmware or any other combination. When implemented in software, the above-described embodiments may be implemented in whole or in part in the form of a computer program product. The computer program product includes one or more computer instructions or computer programs. When the computer instructions or computer programs are loaded or executed on a computer, all or part of the processes or functions described in the embodiments of the present application are generated. The computer may be a general purpose computer, special purpose computer, computer network, or other programmable device. The computer instructions may be stored in or transmitted from one computer readable storage medium to another computer readable storage medium, for example, the computer instructions may be downloaded from a website site, computer, server or data center Transmission by wire or wireless to another website site, computer, server or data center. The computer-readable storage medium may be any available medium that a computer can access, or a data storage device such as a server, a data center, or the like containing one or more sets of available media. The usable media may be magnetic media (eg, floppy disks, hard disks, magnetic tapes), optical media (eg, DVDs), or semiconductor media. The semiconductor medium may be a solid state drive.

It should be understood that, in various embodiments of the present application, the size of the sequence numbers of the above-mentioned processes does not mean the sequence of execution, and the execution sequence of each process should be determined by its functions and internal logic, and should not be dealt with in the embodiments of the present application. implementation constitutes any limitation.

In the several embodiments provided in this application, it should be understood that the disclosed method, apparatus and system may be implemented in other manners. For example, the device embodiments described above are only illustrative; for example, the division of the units is only a logical function division, and there may be other division methods in actual implementation; for example, multiple units or components may be combined or Can be integrated into another system, or some features can be ignored, or not implemented. On the other hand, the shown or discussed mutual coupling or direct coupling or communication connection may be through some interfaces, indirect coupling or communication connection of devices or units, and may be in electrical, mechanical or other forms.

The units described as separate components may or may not be physically separated, and components displayed as units may or may not be physical units, that is, may be located in one place, or may be distributed to multiple network units. Some or all of the units may be selected according to actual needs to achieve the purpose of the solution in this embodiment.

In addition, each functional unit in each embodiment of the present invention may be integrated into one processing unit, or each unit may be physically included individually, or two or more units may be integrated into one unit. The above-mentioned integrated unit may be implemented in the form of hardware, or may be implemented in the form of hardware plus software functional units.

The above-mentioned integrated units implemented in the form of software functional units can be stored in a computer-readable storage medium. The above-mentioned software functional unit is stored in a storage medium, and includes several instructions to cause a computer device (which may be a personal computer, a server, or a network device, etc.) to execute some steps of the methods described in the various embodiments of the present invention. The aforementioned storage medium includes: U disk, mobile hard disk, Read-Only Memory (ROM for short), Random Access Memory (RAM for short), magnetic disk or CD, etc. that can store program codes medium.

It should be understood that the term "and/or" in this document is only an association relationship to describe associated objects, indicating that there can be three kinds of relationships, for example, A and/or B, which can mean that A exists alone, and A and B exist at the same time , there are three cases of B alone. In addition, the character "/" in this text indicates that the related objects are an "or" relationship.

The "plurality" in the embodiments of the present application refers to two or more.

The descriptions of the first, second, etc. appearing in the embodiments of the present application are only used for illustration and distinguishing the description objects, and have no order. any limitations of the examples.

The "connection" in the embodiments of the present application refers to various connection modes such as direct connection or indirect connection, so as to realize communication between devices, which is not limited in the embodiments of the present application.

Although the present invention is disclosed above, the present invention is not limited thereto. Any person skilled in the art can make various changes and modifications without departing from the spirit and scope of the present invention. Therefore, the protection scope of the present invention should be based on the scope defined by the claims.

Claims (14)

1. an energy control method for a composite micro-energy system, characterized in that the composite micro-energy system comprises a micro-energy collection module, an energy storage module and a power supply module, and the micro-energy collection module is used to collect at least one energy, The energy storage module is used for storing the converted electric energy, the power supply module is used for supplying power to a load, and the method includes: Identify the characteristic information of each energy collected by the micro-energy collection module; Inputting the characteristic information of each energy and the electric power required by the load into the decision tree model to obtain a decision label for each energy, where the decision label is used to indicate the direction of each energy; Control the direction of each energy according to the decision label, so as to transmit at least a part of each energy to the power supply module and/or the energy storage module; Wherein, the decision tree model is generated by using a plurality of first energy samples as training data for training. 2 . The energy control method of a composite micro-energy system according to claim 1 , wherein the characteristic information includes one or more of the following: the type of energy and the electric power that the energy can convert. 3 . 3. the energy control method of composite micro-energy system according to claim 1, is characterized in that, the method that obtains the decision label of every kind of energy also comprises: The state of charge of the energy storage module is input into the decision tree model together with the characteristic information of each energy and the electric power required by the load to obtain a decision label for each energy. 4. The energy control method of the composite micro-energy system according to claim 1, wherein the generation method of the decision tree model comprises: Acquiring a plurality of first energy samples, each first energy sample includes electrical power required by the load, at least one sample energy and its characteristic information and decision label; Using the plurality of first energy samples as the training data, the decision tree model is generated by training. 5. The energy control method of the composite micro-energy system according to claim 4, characterized in that, using the plurality of first energy samples as the training data, before training to generate the decision tree model, the method further comprises: include: Determine the range of possible values of the electrical power required by the load and the range of possible values of the characteristic information of each sample energy; The plurality of first energy samples are screened according to the characteristic information of each sample energy in each first energy sample and the electric power required by the load corresponding to the first energy sample, so as to obtain a method for generating the decision tree model. first energy sample. 6. The energy control method of the composite micro-energy system according to claim 1, wherein before inputting the characteristic information of each energy and the electric power required by the load into the decision tree model, the method further comprises: Acquire a plurality of second energy samples, each second energy sample includes the electrical power required by the load, at least one sample energy and its characteristic information and a preset decision label, wherein the plurality of second energy samples and the The plurality of first energy samples are independent of each other; Input the electric power required by the load in each second energy sample and the characteristic information of each kind of sample energy into the decision tree model, to obtain the model decision label of each kind of sample energy in the second energy sample; Compare the model decision label of each kind of sample energy in each second energy sample with the preset decision label, and calculate the second energy that the model decision label of each sample energy is consistent with the preset decision label The number of samples accounts for the proportion of the number of all second energy samples. If the proportion is smaller than the first preset threshold, the decision tree model is pruned to update the decision tree model. 7 . The energy control method of a composite micro-energy system according to claim 1 , wherein the energy storage module comprises a plurality of energy storage elements, and the energy storage module is further used to supply power to the load, and the energy storage module Methods also include: Obtain the state of charge of multiple energy storage elements; According to the electric power to be provided by the energy storage module and the state of charge of the plurality of energy storage elements, a fuzzy control algorithm is used to obtain the power supply ratio of the plurality of energy storage elements, wherein the electric power to be provided by the energy storage module is The difference between the electric power required by the load and the energy-convertible electric power of the micro-energy collection module; The output electric power of each energy storage element is determined according to the power supply ratio and the electric power to be provided by the energy storage module. 8 . The energy control method of a composite micro-energy system according to claim 7 , wherein the plurality of energy storage elements comprise supercapacitors. 9 . 9 . The energy control method of the composite micro-energy system according to claim 7 , wherein, according to the electric power to be provided by the energy storage module and the state of charge of the plurality of energy storage elements, a fuzzy control algorithm is used to obtain the energy control method. 10 . The power supply ratio of multiple energy storage elements includes: The power range to which the electric power to be provided by the energy storage module belongs is determined according to the electric power to be provided by the energy storage module, and the charge range to which the state of charge of each energy storage element belongs is determined according to the state of charge of each energy storage element ; Query the fuzzy control rule table according to the power range and the charge range of each energy storage element to determine the power supply ratio range of each energy storage element; Defuzzify the power supply ratio range of each energy storage element to obtain the power supply ratio of each energy storage element; Wherein, the fuzzy control rule table is used to describe the mapping relationship between the power range and the charge range and the power supply ratio range. 10. The energy control method of the composite micro-energy system according to claim 1, wherein the micro-energy collection module comprises a plurality of micro-energy collectors, and the method further comprises: acquiring first feedback information, where the first feedback information is used to indicate whether the micro-energy collector is replaced; Determine whether the micro-energy collector is replaced according to the first feedback information, and if so, obtain the energy convertible electric power of each replaced micro-energy collector; Compare the energy convertible electrical power of each replaced micro-energy collector with the second preset threshold, and if the energy convertible electrical power of any replaced micro-energy collector does not exceed the second preset threshold, retrain to generate The decision tree model. 11. The energy control method of a composite micro-energy system according to claim 1, wherein the energy storage module comprises a plurality of energy storage elements, and the method further comprises: acquiring second feedback information, where the second feedback information is used to indicate whether the energy storage element is replaced; Determine whether the energy storage element is replaced according to the second feedback information, and if so, acquire the state of charge of each energy storage element after replacement; comparing the upper limit of the state of charge of each energy storage element after replacement with a third preset threshold; If the upper limit of the state of charge of any replaced energy storage element does not exceed the third preset threshold, the decision tree model is retrained to generate the decision tree model. 12. An energy control device for a composite micro-energy system, wherein the composite micro-energy system comprises a micro-energy collection module, an energy storage module and a power supply module, and the micro-energy collection module is used to collect at least one energy, The energy storage module is used to store the converted electrical energy, the power supply module is used to supply power to the load, and the device includes: an identification module for identifying the characteristic information of each energy collected by the micro-energy collection module; a classification module, configured to input the characteristic information of each energy and the electric power required by the load into the decision tree model to obtain a decision label for each energy, and the decision label is used to indicate the trend of each energy; a transmission control module, configured to control the direction of each energy according to the decision label, so as to transmit at least a part of each energy to the power supply module and/or the energy storage module; Wherein, the decision tree model is generated by using a plurality of first energy samples as training data for training. 13. A storage medium on which a computer program is stored, wherein when the computer program is run by a processor, the energy control method of the composite micro-energy system according to any one of claims 1 to 11 is executed. step. 14. A composite micro-energy system, characterized in that the system comprises: a micro-energy collection module for collecting at least one type of energy; an energy storage module for storing the converted electrical energy; The power supply module is used to supply power to the load; The controller is used for executing the steps of the energy control method of the composite micro-energy system according to any one of claims 1 to 11.

Images