CN106979641B - Based on the refrigeration system data driving energy-saving control system and method for improving MFAC - Google Patents

Abstract

The invention discloses a refrigeration system data-driven energy-saving control system based on improved MFAC, which includes a constant frozen water supply temperature control loop composed of a PID controller and a frequency converter, and a model-free adaptive algorithm controller, an electronic expansion valve and an evaporator. Constitute the variable minimum superheat control loop. According to the change of the system load, adjust the frequency of the compressor through the PID controller to keep the temperature of the chilled water supply constant to realize the matching of the cooling capacity and the heat load; use the experimental method to obtain the relationship curve between the system load and the minimum stable superheat of the evaporator. Real-time calculation system According to the relationship curve, the minimum stable superheat corresponding to the load is calculated by the linear difference method; in the variable minimum superheat control loop, the minimum stable superheat corresponding to the refrigeration load is used as the superheat setting value, The MFAC control algorithm with a lag time constraint is used to control the superheat of the evaporator through an electronic expansion valve.

Description

Refrigeration system data-driven energy-saving control system and method based on improved MFAC

technical field

The invention relates to an energy-saving control method of an MFAC refrigeration system with a lag time constraint, and belongs to the field of optimal control of an air-conditioning system.

Background technique

With the continuous acceleration of my country's urbanization process, a large number of office buildings, hotels, shopping malls and other buildings have been built, and these buildings are equipped with central air-conditioning systems. With the improvement of people's living standards, air-conditioning facilities have been used in urban residential buildings and even houses in villages and towns. At present, my country's building energy consumption accounts for about 30% of the total energy consumption, of which air-conditioning energy consumption accounts for 50% to 70%, and air-conditioning energy consumption even accounts for 1/3 of the total urban electricity load during the peak electricity consumption period in summer. Most of my country's air-conditioning facilities are designed according to the maximum cooling and heating load, but most of the time the air-conditioning works under partial load, and there is a lot of room for energy saving. Air-conditioning energy consumption is the main body of building energy consumption, and air-conditioning energy saving is an important task for China to achieve sustainable development ^[1] .

At present, there are three commonly used refrigeration methods, namely, vapor compression, steam injection and absorption refrigeration. The above refrigeration methods directly consume electric energy or heat energy. According to the analysis of the theoretical refrigeration cycle and actual energy consumption, it can be concluded that the compression refrigeration method has a high cooling capacity per unit energy consumption, but there is a problem of low partial load energy efficiency. The primary task of refrigeration system control is to ensure the performance indicators required by the process of the refrigeration system through appropriate control functions when the load and external conditions change, and to keep the system running within a safe and reasonable working condition range. The further task is to Improve the operating economy of the system under various working conditions as much as possible ^[2] . The refrigeration control technology represented by the electronic expansion valve and compressor frequency conversion control fully penetrates the control means into the refrigeration cycle, so that the entire refrigeration system can achieve the purpose of economical and efficient operation.

The compression refrigeration system is a nonlinear system with multiple disturbances, variable parameters, strong coupling, multiple working conditions, and large inertia time delay. During the operation of the refrigeration system, changes in the external ambient temperature, light conditions, the number of people in the building, and the heat generated by the electrical equipment in the building lead to changes in the parameters and load requirements of the refrigeration system. It is necessary to study the control variables of the refrigeration system In order to determine the influence of disturbance parameter changes on the refrigeration cycle and determine the optimal control scheme, it is necessary to establish a nonlinear dynamic mathematical model suitable for large-scale changes and with time-varying parameters. However, since the refrigerant has both single-phase and gas-liquid phases in the heat exchanger, the modeling of the heat exchanger is very complicated, and there are various difficult modeling factors, making it difficult to establish a refrigeration system model, even if it is successfully established , the model will also have complex, high-order difficult to use in advanced controller design or low precision difficult to reflect the internal dynamics of the refrigeration system, so a control algorithm that does not depend on the model and has a better control effect is needed ^[3] .

At present, in practical applications, PID and its improved algorithm are still mostly used to design controllers. Although most systems can achieve stable operation under PID control, because the air conditioning system is a coupling system with large inertia and large lag ^[4] , there are still some deficiencies. First of all, the PID algorithm is very effective in the control of linear constant systems, but the central air-conditioning system varies with the system load and the external environment, and the system model parameters are time-varying, so the control effect is not ideal. Secondly, the parameter setting of the PID controller is related to the system model, but when the same parameters are applied to a new system model, the performance of the system will become worse or even unstable. Finally, there is a large lag problem in the actual refrigeration system, and the existing control methods based on PID and the like are difficult to achieve effective control of the large time-delay system.

It can be seen from the above analysis that the optimal control method for the air-conditioning and refrigeration system still needs to be further improved and innovated. In view of the characteristics of multiple inputs and multiple outputs of the actual refrigeration system, strong coupling, large lag, and difficulty in establishing an accurate mathematical model that can realize control, it is necessary to invent a new type of cooling system that has a short control time, less dependence on system model information, and a relatively simple algorithm that is easy to implement. The control algorithm is used to effectively realize the energy-saving optimal control of the air-conditioning and refrigeration system.

[references]

[1] Sun Yulei; Discussion on Building Environment and HVAC Energy Conservation [J]; Science and Technology Innovation and Application, 2016,19(7):272-272.

[2] Zhai Wenpeng, Wu Aiguo, You Yuwen, et al. Research on Generalized Predictive Control of Refrigeration System [J]; Cryogenics and Superconductivity, 2012(2):28-33.

[3] Ahamed J U, Saidur R, Masjuki H H. A review on exergy analysis of vapor compression refrigeration system [J]. Renewable & Sustainable Energy Reviews, 2011, 15(3): 1593-1600.

[4] Tian J, Feng Q, Zhu R. Analysis and experimental study of MIMO control in refrigeration system [J]. Energy Conversion & Management, 2008, 49(5): 933-939.

Contents of the invention

The purpose of the present invention is to solve the problems existing in the prior art, and propose a refrigeration system data-driven energy-saving control system and method based on the improved MFAC. This method adds the input change rate constraint item with lag time into the control rate calculation process, which can better solve the large time lag problem and greatly improve the response speed of the refrigeration system; and this method is based on the controlled system It does not need to establish a mathematical model of the controlled system, the amount of calculation is small, it is easy to implement, and it can avoid problems such as poor control effect that may be caused by inaccurate system models.

In order to solve the above technical problems, the present invention proposes a data-driven energy-saving control system for refrigeration systems based on improved MFAC, including a PID controller and a frequency converter connected to the controlled refrigeration system, and a condenser and an inverter connected to the controlled refrigeration system. An expansion valve between the evaporators, the expansion valve is connected with a model adaptive algorithm controller, and the expansion valve is an electronic expansion valve; the PID controller and the frequency converter form a constant frozen water supply temperature control loop; The model-free adaptive algorithm controller, the electronic expansion valve and the evaporator form a variable minimum superheat control loop.

The control method for driving the energy-saving control system based on the data of the refrigeration system based on the improved MFAC is to control the degree of superheat of the evaporator in real time by controlling the opening degree of the electronic expansion valve, and make the degree of superheat track the upper minimum stable degree of superheat ,Proceed as follows:

Step 1. Obtain the set value of the superheat of the evaporator:

Firstly, the Qy ^* curve of the relationship between the system load of the controlled cooling system and the minimum stable superheat degree of the evaporator is obtained through experiments; then, the constant chilled water supply temperature control loop is used to calculate the system at the current moment according to formula (1) Load Q; finally, according to the above Qy ^* curve, and through the linear difference method, the minimum stable superheat corresponding to the system load at the current moment is the evaporator superheat setting value y ^* ;

Q＝CM△T (1)

In formula (1), C is the specific heat capacity coefficient of water, M is the chilled water flow rate, and △T is the temperature difference between supply and return water;

Step 2. Energy-saving control based on MFAC with lag time constraints, including:

2-1) Input items of the model-free adaptive algorithm controller, including:

y(k), y(k-1), u(k-1), u(k-2), u(k-1-τ), u(k-2-τ), y ^* (k+1 ); where, y ^* is the set value of the superheat degree obtained in step 1, y is the actual superheat degree fed back by the evaporator, u is the opening degree of the electronic expansion valve; y(k) is the actual superheat degree of the evaporator at time k , u(k) is the opening degree of the electronic expansion valve at time k, y ^* (k+1) is the set value of the superheat degree of the evaporator at time k+1, τ is the lag time constant used to improve the system response speed, 2 -2) The output of the model-free adaptive algorithm controller is the opening u(k) of the electronic expansion valve at time k,

In formula (2), T is the sampling time, ρ, η, and λ are all weight coefficients, ρ∈(0,2), η∈(0,10), λ∈(0,100), and τ can be controlled according to The ratio of the actual lag time to the sampling time is given. Generally, when the sampling time T is 0.1s, τ∈[100,200],

In formula (3), ξ, μ are weight coefficients; ξ∈(0,2), μ∈(0,10),

2-3) Loop control of model-free adaptive algorithm controller:

From formula (3) through the opening of the electronic expansion valve at k-1 time and k-2-τ time, the actual superheat of the evaporator at k-1 time and k-1 time, and the characteristic parameters at k-1 time, the k time The characteristic parameters of From formula (2), through the opening of the electronic expansion valve at k-1 time and k-1-τ time, the actual superheat degree of the evaporator at k time, the characteristic parameter at k time, the expected output at k+1 time, that is, the evaporator superheat The heat setting value is calculated to obtain the control input of the controlled cold system at this moment, that is, the opening degree u(k) of the electronic expansion valve at the current moment, so as to obtain the output of the controlled cold system, that is, the actual superheat degree y(k) of the evaporator +1), the obtained data is used as the input data of the model-free adaptive algorithm controller in the next cycle of the variable minimum superheat control loop.

Compared with prior art, the beneficial effect of the present invention is:

First, the electronic expansion valve is used to control the superheat of the evaporator, the adjustment is rapid and stable, and the control effect is more ideal; and the method of changing the set value of superheat is adopted, compared with the conventional control method of fixed superheat set value, because The coupling effect of the adjustment process of the compressor and the expansion valve is reduced, the dynamic adjustment process of the compressor and the electronic expansion valve tends to be smooth, the dynamic stability of the system is increased, and the efficiency of the evaporator is increased when the variable superheat control is adopted , improve the cooling capacity and performance coefficient of the system.

Second, in view of the fact that the device models of refrigeration system evaporators are too complex and modeling is relatively difficult, the present invention uses a control algorithm based on improved MFAC to realize the control of the superheat of the evaporator by the electronic expansion valve, effectively solving the problem of traditional control The dependence of the algorithm on the system model, while avoiding the problem of poor control performance caused by inaccurate system models, enhances the adaptability to working conditions and environmental changes, and makes the control effect better.

Third, aiming at the large lag characteristics of the refrigeration system, the present invention, on the basis of the basic MFAC algorithm, adds a constraint term of the input change rate with a lag time in the control input criterion function and the pseudo partial derivative estimation criterion function, thus obtaining The MFAC control algorithm with lag time constraints is used to control the variable superheat loop, reduce the response time in the process of electronic expansion valve controlling the evaporator, and greatly improve the response speed of the refrigeration system.

Description of drawings

Fig. 1 is the schematic diagram of the energy-saving control scheme of the refrigeration system in the present invention;

Fig. 2 is an overall control structure diagram of the refrigeration system in the present invention.

Detailed ways

The technical solution of the present invention will be further described in detail below in conjunction with the accompanying drawings and specific embodiments, and the described specific embodiments are only for explaining the present invention, and are not intended to limit the present invention.

The design idea of the present invention is: Aiming at the large lag characteristics of the air-conditioning and refrigeration system, the present invention designs a model-free adaptive control method with lag time input change rate constraints, so as to reduce the gap between the electronic expansion valve and the evaporator. The problem of large time lag can effectively improve the response speed of the refrigeration system; and the present invention adopts the control scheme of variable superheat setting value, that is, the control method of finding the corresponding minimum stable superheat degree as the set value with the change of load, reducing The coupling effect of the adjustment process of the compressor and the expansion valve is reduced, the dynamic stability of the refrigeration system is enhanced, and the efficiency of the evaporator is increased at the same time, so that the cooling capacity and COP (coefficient of performance) of the system are both improved to a certain extent.

As shown in Figure 1, a data-driven energy-saving control system for refrigeration systems based on improved MFAC proposed by the present invention, the controlled refrigeration system is shown in dotted lines in Figure 1, including evaporators, compressors and condensers, and the control system includes A PID controller and a frequency converter connected to the controlled cold system, and an expansion valve connected between the condenser and the evaporator in the controlled cold system, the expansion valve is connected with a model-free adaptive algorithm controller, the expansion The valve is an electronic expansion valve; as shown in Figure 2, the PID controller and the frequency converter constitute a constant chilled water supply temperature control loop; the model-free adaptive algorithm controller, the electronic expansion valve and the evaporation The controller constitutes a variable minimum superheat control loop, thereby forming a double closed-loop control scheme. The constant chilled water supply temperature control loop adjusts the frequency of the compressor according to the change of the system load to keep the chilled water supply temperature constant, so as to realize the matching of cooling capacity and heat load; The variable minimum superheat degree control loop makes the superheat degree of the evaporator follow a given value by adjusting the electronic expansion valve. In the constant chilled water supply temperature control loop, the PID control algorithm can effectively control the cooling capacity of the compressor, so this loop still uses the PID control algorithm. In the variable minimum superheat control loop, liquid-gas conversion and other processes are required in the evaporator, and there is a large hysteresis, and the model of the evaporator and other devices is too complicated, and it is difficult to establish a model that can achieve precise control. A model-free adaptive control algorithm with lag time constraints is used to achieve fast and effective control of the superheat of the evaporator.

The control method for driving the energy-saving control system based on the data of the improved MFAC refrigeration system is mainly to control the degree of superheat of the evaporator in real time by controlling the opening of the electronic expansion valve. The present invention adopts variable minimum degree of superheat control, namely Use the expansion valve to control the superheat of the evaporator in real time, so that the superheat can track the minimum stable superheat. The smaller the expansion valve opening, the greater the superheat; otherwise, the smaller the superheat.

Specific steps are as follows:

1. Calculate the superheat setting value of the evaporator:

First, the Qy ^* curve of the relationship between the system load of the controlled cooling system and the minimum stable superheat degree of the evaporator is obtained through experiments. The heat load of the evaporator determines the minimum superheat degree at which the controlled system can operate stably. The set value of the superheat degree in the variable superheat degree circuit needs to be given according to the minimum stable superheat degree curve of the evaporator. Therefore, it is necessary to obtain the system load and Dependency curve of minimum stable signal (Qy ^* line). The minimum stable superheat of the evaporator regulated by the electronic expansion valve under various loads can be obtained by means of experiments. The experimental steps are as follows:

1) Real-time detection of cooling system water supply temperature (chilled water supply water temperature), return water temperature (chilled water return water temperature) and chilled water pump flow, and calculate real-time load values.

2) Adjust the frequency of the compressor as a means of adjusting the refrigeration load. The adjustment range is from 50Hz to 30Hz, the adjustment step is 2Hz, and the adjustment time interval is based on the first step. The real-time refrigeration load calculation value stabilizes as the adjustment time node.

3) Set a higher value for the initial set value of the electronic expansion valve control superheat circuit, generally above 10 degrees, and gradually reduce the overheating value with a step size of 0.1°C under a certain fixed compressor frequency in the second step. Heat setting value, and observe the adjustment of the superheat degree of the output value of the control loop. As the heat setting value gradually decreases, the fluctuation range of the actual superheat degree gradually increases; when the fluctuation range of the superheat degree reaches ±0.4°C, it is considered that the state parameters of the evaporator (such as the evaporating pressure) have a fixed amplitude automatic fluctuation. excited oscillation. Add 0.1°C to the set value of superheat at this moment, which is the minimum stable superheat under the load.

4) According to the above method, the minimum stable superheat of the system under various refrigeration capacity conditions is obtained, and fitted into a Qy ^* line.

Then, use the constant chilled water supply temperature control loop, and calculate the system load Q at the current moment according to formula (1); finally, according to the above Qy ^* curve, and use the linear difference method to obtain the minimum stable process load corresponding to the system load at the current moment The heat is the evaporator superheat setting value y ^* ;

Q＝CM△T (1)

In formula (1), C is the specific heat capacity coefficient of water, M is the chilled water flow rate, and ΔT is the temperature difference between supply and return water.

Second, energy-saving control based on MFAC with lag time constraints.

Now it is widely used to control the electronic expansion valve by PID or PI algorithm to adjust the superheat of the evaporator, but there are some problems. There is a big hysteresis problem in the actual refrigeration system, and the existing control methods based on PID and the like are difficult to achieve effective control of the large time-delay system; PID parameter tuning needs to be based on a simplified and unchanged model, but The mathematical model of the superheater system is complex and is easily affected by conditions such as load and operating conditions, and the control effect is not ideal. Therefore, the present invention designs a data-driven control method for the superheat of the electronic expansion valve, that is, a novel MFAC control algorithm based on a lag time constraint. Under variable load conditions, the superheat of the evaporator can track the given value stably and quickly by adjusting the opening of the expansion valve through the MFAC controller with a lag time constraint.

The control algorithm based on MFAC is not ideal for the object with large time-delay like the refrigeration system, and it is difficult to effectively fast track the control. According to the characteristic that the controlled object has a large time lag, the invention designs a model-free self-adaptive control method with lag time input change rate constraints. In the control input criterion function, the rate of change between two sets of input values with an interval lag time τ is selected as an important constraint parameter in the input criterion, which is called the constraint item of the input change rate with lag time. The present invention considers the influence of the delay time constant τ on the large time-delay system. In order to better control the objects with large time-delay, on the basis of the basic MFAC algorithm, especially for the large time-delay system, the algorithm with the delay time τ is added. Input the constraint item of change rate, optimize the MFAC control algorithm, and greatly improve the response speed of the air conditioner.

Based on the control algorithm of MFAC, each iteration in the operation process only needs the measurement data of the existing closed-loop experiment to obtain the characteristic parameter φ(k), and then generate the control signal u(k). The whole control process does not need the model of the controlled object information.

The specific control strategy of the MFAC algorithm with lag time constraints is as follows:

Model-free adaptive control starts from the input and output data of the controlled cooling system, and obtains the next control input signal from the known input and output data.

{[u(k-1),y(k)]} and {[u(k),y(k+1)]} are observation data at adjacent sampling moments, and u(k)≠u(k- 1), where u(k) and y(k) are the input and output of the system respectively. As a model-free control algorithm for controllers, its input items include:

y(k), y(k-1), u(k-1), u(k-2), u(k-1-τ), u(k-2-τ), y ^* (k+1 );

Among them, y ^* is the set value of the superheat degree obtained in step 1, y is the actual superheat degree fed back by the evaporator, u is the opening degree of the electronic expansion valve; y(k) is the actual superheat degree of the evaporator at time k, u (k) is the opening degree of the electronic expansion valve at time k, y ^* (k+1) is the set value of the superheat degree of the evaporator at time k+1, τ is the lag time constant used to improve the system response speed,

y(k) is the output of the controlled object at time k, that is, the superheat degree of the evaporator at this time; u(k) is the control input of the controlled object at time k, that is, the opening degree of the electronic expansion valve; y ^* (k+ 1) is the expected output of the controlled object at time k+1, which is the set value of the superheat of the evaporator; τ is the lag time constant added to improve the system response speed.

The output of the model-free adaptive algorithm controller (that is, the control input of the system) is the opening u(k) of the electronic expansion valve at time k, which can be obtained by formula (2):

Given the initial data, the available data include the superheat of the evaporator {y(k), y(k-1)}, the opening of the electronic expansion valve {u(k-1), u(k-2), u(k-2-τ)}, where the k value represents the k moment.

In formula (2), T is the sampling time, ρ, η, and λ are all weight coefficients, ρ∈(0,2), η∈(0,10), λ∈(0,100), and τ can be controlled according to The ratio of the actual lag time to the sampling time is given. Generally, when the sampling time T is 0.1s, τ∈[100,200], only the characteristic parameters in formula (2) unknown, parameter estimation of pseudo partial derivatives is required, when Or when △u(k-1)≤ε (ε is a sufficiently small positive number), the characteristic parameter In other cases, the characteristic parameters Obtained by formula (3).

In formula (3), ξ, μ are weight coefficients, ξ∈(0,2), μ∈(0,10), u(k-1)-u(k-2) are the adjacent moments of expansion valve opening The difference, y(k)-y(k-1) is the difference between adjacent moments of superheat of the evaporator, the characteristic parameter It is an important parameter to estimate the opening degree u(k) of the electronic expansion valve.

Loop control of the model-free adaptive algorithm controller: the opening degree of the electronic expansion valve at time k-1 and k-2-τ, the actual superheat of the evaporator at time k and k-1, k The characteristic parameter at time -1 is obtained to obtain the characteristic parameter at time k From formula (2), through the opening of the electronic expansion valve at time k-1 and k-1-τ, the actual superheat degree of the evaporator at time k, the characteristic parameters at time k, and the expected output at time k+1, that is, the evaporator superheat The heat setting value is calculated to obtain the control input of the controlled cold system at this moment, that is, the opening degree u(k) of the electronic expansion valve at the current moment, so as to obtain the output of the controlled cold system, that is, the actual superheat of the evaporator y(k +1), the y ^* (k+1) is the minimum stable superheat corresponding to the refrigeration load, and the superheat of the evaporator is controlled by controlling the opening of the expansion valve, so that it can quickly and stably track the upper given value. The output of the system {y(k), y(k-1)}, the control input of the system {u(k-1), u(k-2), u(k-1-τ), u(k- 2-τ)} and the set value of superheat y ^* (k+1) are used as the input of the model-free adaptive controller with lag time constraints, and the characteristic parameters of the values are continuously updated As a connection, the control signal u(k) is calculated, that is, the opening degree of the electronic expansion valve, and the signal is fed back to the air-conditioning and refrigeration system to complete the closed-loop control.

To sum up, the steps implemented by the control method of the present invention can be summarized as follows: First, in the constant chilled water supply temperature control loop, the PID controller adjusts the frequency of the compressor according to the change of the system load to keep the chilled water supply temperature constant and realize the cooling capacity Match the heat load; then use experimental means to obtain the relationship curve between the system load and the minimum stable superheat of the evaporator, calculate the system load in real time, and calculate the minimum corresponding to the load by the method of linear difference according to this relationship curve Stable superheat; finally, in the variable minimum superheat control loop, the minimum stable superheat corresponding to the refrigeration load is used as the superheat setting value, and the MFAC control algorithm with a lag time constraint is used to control the evaporator through the electronic expansion valve. superheat.

In the constant chilled water supply temperature control of the present invention, the product of the temperature difference between the supply and return water of the chilled water system in the air conditioning system and the flow rate is proportional to the system load, which reflects the cooling capacity of the actual demand of the system, and the frequency of the compressor in the chiller system The higher the value, the greater the cooling capacity. Therefore, adjust the operating frequency of the inverter compressor according to the actual cooling capacity required by the system, thereby adjusting the cooling capacity of the system, keeping the temperature of the chilled water supply constant, and matching the cooling capacity with the heat load.

In addition to the traditional frequency conversion energy saving of the compressor, the present invention also designs the link of variable minimum superheat control to achieve better energy saving effect. The frequency of the compressor changes continuously according to the change of the system load, which changes the cooling capacity of the refrigeration unit. The setting value of the variable superheat circuit is set according to the minimum stable superheat curve, which will change with the change of refrigeration load. At this time, it is necessary to continuously adjust the expansion valve dynamically, so that the superheat of the evaporator can track set value.

Although the present invention has been described above in conjunction with the accompanying drawings, the present invention is not limited to the above-mentioned specific embodiments, and the above-mentioned specific embodiments are only illustrative, rather than restrictive. Under the enlightenment of the present invention, many modifications can be made without departing from the gist of the present invention, and these all belong to the protection of the present invention.

Claims (1)

1. a kind of driving energy-saving control method based on the refrigeration system data for improving MFAC, involved based on the system for improving MFAC Cooling system data-driven energy-saving control system, includes the PID controller being connect with controlled refrigeration system and frequency converter, and be connected to Expansion valve in controlled refrigeration system between condenser and evaporator, the expansion valve are connected with the control of model-free adaption algorithm Device, the expansion valve are electric expansion valve；The PID controller and the frequency converter constitute a constant freezing Water temperature control Circuit；The model-free adaption algorithmic controller, the electric expansion valve and the evaporator constitute one and become minimum superheat Control loop；It is characterized in that, driving energy-saving control system based on the refrigeration system data for improving MFAC using above-mentioned, pass through control The degree of superheat of evaporator described in the aperture real-time control of the electric expansion valve is made, and the upper minimum of degree of superheat tracking is made to stablize Temperature, steps are as follows：

Step 1: seeking evaporator superheat setting value：

First, the pass being controlled between the system loading and evaporator minimum thermal stability degree of refrigeration system is obtained by experimental method It is Q-y^*Curve；Then, using constant freezing Water temperature control circuit, and the system that current time is calculated according to formula (1) Load Q；Finally, according to above-mentioned Q-y^*Curve, and the minimum corresponding to current time system loading is obtained by linear difference method Thermal stability degree is evaporator superheat setting value y^*；

Q=CM Δs T (1)

In formula (1), C is the specific heat capacity coefficient of water, and M is chilled-water flow, and Δ T is supply backwater temperature difference；

Step 2: based on the MFAC Energy Saving Controls constrained with lag time, including：

2-1) the input item of model-free adaption algorithmic controller, including：

Y (k), y (k-1), u (k-1), u (k-2), u (k-1- τ), u (k-2- τ), y^*(k+1)；Wherein, y^*It is obtained for step 1 The setting value of the degree of superheat, y are the practical degree of superheat of evaporator feedback, and u is the aperture of electric expansion valve；Y (k) is to evaporate at the k moment The practical degree of superheat of device, u (k) are the aperture of moment k electric expansion valve, y^*(k+1) it is that k+1 moment evaporator superheats are set Value, τ be for the lag time constant for improving system response time,

2-2) output of model-free adaption algorithmic controller is the aperture u (k) of moment k electric expansion valve,

In formula (2), T is the sampling time, ρ, η, and λ is weight coefficient, ρ ∈ (0,2), η ∈ (0,10), λ ∈ (0,100), τ according to The ratio in controlled refrigeration system actual hysteretic time and sampling time provides, when the sampling time, T took 0.1s, τ ∈ [100, 200],

In formula (3), ξ, μ are weight coefficient, ξ ∈ (0,2), μ ∈ (0,10)；

2-3) the loop control of model-free adaption algorithmic controller：

It is practical by the aperture of the electric expansion valve at k-1 moment and k-2- τ moment, k moment and k-1 moment evaporator by formula (3) The degree of superheat, the characteristic parameter at k-1 moment, obtain the characteristic parameter at k momentWhen by formula (2) by k-1 moment and k-1- τ The aperture of the electric expansion valve at quarter, the practical degree of superheat of k moment evaporators, the characteristic parameter at k moment, the desired output at k+1 moment That is the control input i.e. electronic expansion at current time of the moment controlled refrigeration system is calculated in evaporator superheat setting value The aperture u (k) of valve, to obtain the output of controlled refrigeration system, the i.e. practical degree of superheat y (k+1) of evaporator, obtained data As the input data for becoming minimum superheat control loop subsequent cycle process model-free adaption algorithmic controller.