WO2022179008A1

WO2022179008A1 - Supply chain finance ai daas algorithm warehouse platform based on blockchain

Info

Publication number: WO2022179008A1
Application number: PCT/CN2021/101105
Authority: WO
Inventors: 刘天琼
Original assignee: 深圳市爱云信息科技有限公司
Priority date: 2021-02-24
Filing date: 2021-06-19
Publication date: 2022-09-01
Also published as: CN112883116A

Abstract

A supply chain finance AI DaaS algorithm warehouse platform based on a blockchain, the supply chain finance AI DaaS algorithm warehouse platform based on a blockchain comprising an application layer, an open interface layer, an algorithm warehouse service layer, a data layer and an infrastructure layer, which are connected in sequence, wherein the application layer is used for providing a supply chain finance service based on a blockchain; the open interface layer is used for providing, to the application layer, a service for connecting to the algorithm warehouse service layer; the data layer is used for providing data service support to the algorithm warehouse service layer; and the infrastructure layer is used for providing development support to the data layer. The problems of algorithm fusion and sharing of supply chain finance based on a blockchain can be solved.

Description

Blockchain-based supply chain finance AI DaaS algorithm warehouse platform

technical field

The invention relates to the field of supply chain finance of blockchain, in particular to a supply chain finance AI DaaS algorithm warehouse platform based on blockchain.

Background technique

DaaS (Data as a Service) is an efficient way to provide and manage multiple powerful desktop configurations at a predictable per-user cost. The flexibility and agility it brings enables remote workers, regular and temporary employees, and even users with multiple PCs, to get the access and applications they need, no matter where they are.

Because DaaS provides a new way for enterprises to share data with other enterprises through centralized management of data resources and scene-based data; in today's era of data explosion, no enterprise can collect With the DaaS service, you can purchase all the data you need from other companies, and improve the competitiveness of the enterprise through division of labor and collaboration.

With the development and popularization of AI algorithms, since the existing blockchain supply chain finance algorithms cannot be integrated and shared, the development of supply chain finance is seriously hindered. Therefore, a blockchain-based supply chain is urgently needed. The financial AI DaaS algorithm warehouse platform solves the problem of algorithm integration and sharing of blockchain supply chain finance.

SUMMARY OF THE INVENTION

The purpose of the present invention is to provide a blockchain-based supply chain finance AI DaaS algorithm warehouse platform to solve the problem of algorithm fusion and sharing of blockchain-based supply chain finance.

In order to achieve the above purpose, the technical solution adopted in the present invention is as follows: the present invention provides a supply chain finance AI DaaS algorithm warehouse platform based on blockchain, including an application layer, an open interface layer, an algorithm warehouse service layer, A data layer and an infrastructure layer, the application layer is used to provide blockchain-based supply chain financial services, the open interface layer is used to provide the application layer with a service connecting the algorithm warehouse service layer, the data The layer is used to provide data service support for the algorithm warehouse service layer, and the infrastructure layer is used to provide development support for the data layer;

The application layer includes a letter of credit module, a non-recourse financing module, a supply chain finance module, a smart retail module, a link management module, a smart contract script module, a smart contract template module, and a smart contract management platform. It is used to provide a letter of credit business function, the non-recourse financing module is used to provide a non-recourse financing business function, the supply chain financial module is used to provide support services for trusted credit transfer, and the smart retail module is used for In order to provide smart retail business functions, the link management module is used to provide data link management functions, the smart contract script module is used to provide the automatic writing function of smart contract scripts, and the smart contract template module is used to provide intelligent The function of the contract, the smart contract management platform is used to provide the management function of the smart contract;

The open interface layer includes a chain code protocol interface, a blockchain API service interface, an identity authentication service interface, and a light client interface. The chain code protocol interface is a blockchain access interface, and the blockchain API service interface uses In order to provide all the functions related to the blockchain so that the application system only needs to call the relevant interface to become a blockchain application, the identity authentication service interface is used to provide the function of verifying user information, and the light client interface is used to Provide users with a secure and decentralized way to access and interact with the blockchain without synchronizing the entire blockchain;

The algorithm warehouse service layer includes a data visualization algorithm module, a KYC verification algorithm module, a trusted identity system algorithm module, a consensus mechanism algorithm module, a privacy protection algorithm module, and a ledger maintenance algorithm module, and the data visualization algorithm module is used for data visualization algorithms. The KYC verification algorithm module is used for the storage and application of the KYC verification algorithm, the trusted identity system algorithm module is used for the storage and application of the trusted identity system algorithm, and the consensus mechanism algorithm module is used for consensus The storage and application of the mechanism algorithm, the privacy protection algorithm module is used for the storage and application of the privacy protection algorithm, and the account book maintenance algorithm module is used for the storage and application of the account book maintenance algorithm;

The data layer includes a block data sharing module, a block data governance module, and a block data development module. The block data sharing module is used to provide data sharing functions for different modules, and the block data governance module is used to provide different modules. Data governance function, the block data development module is used to provide data development function for different modules;

The infrastructure layer includes Docker, Docker-compose, K8S, the Docker is an open source application container engine, the Docker-compose is an application tool for defining and running multi-container Docker, and the K8S is a container-based cluster management The platform is used to manage Docker and containers.

Further, the data visualization algorithm includes a parallel visualization algorithm, an in-situ visualization algorithm, and a time series data visualization algorithm;

The parallel visualization algorithm includes a task parallel algorithm, a pipeline parallel algorithm, and a data parallel algorithm; the task parallel algorithm divides the visualization process into independent subtasks, and there is no data dependency between the running subtasks; the pipeline The parallel algorithm uses streaming to read data fragments, divides the visualization process into multiple stages, and executes each stage in parallel by the computer to speed up the processing process; the data parallel algorithm is a single-program multiple-data method, which divides the data into multiple subsets, and then use the subset as the granularity to execute the program in parallel to process the different subsets;

The in-situ visualization algorithm is to generate visualization in the process of numerical simulation and is used to alleviate the output bottleneck of large-scale numerical simulation, including image in-situ visualization algorithm, distributed data in-situ visualization algorithm, compressed data in-situ visualization algorithm and feature in-situ visualization. The in-situ visualization algorithm of the image is to map the data into visualization during the numerical simulation process and save it as an image; the in-situ visualization algorithm of the distribution data is calculated according to the statistical index defined by the user during the numerical simulation Statistical indicators are saved, followed by statistical data visualization; the in-situ visualization of the compressed data adopts a compression algorithm to reduce the output scale of the numerical simulation data, and the compressed data is used as the input of the subsequent visualization processing; the feature in-situ visualization algorithm is in the numerical simulation process. Extract features and save them, and use feature data as input for subsequent visualization processing;

The time series data visualization algorithm is used to establish a prediction model, perform predictive analysis and user behavior analysis, including an area chart algorithm, a bubble chart algorithm, a Gantt chart algorithm, a heat map algorithm, a histogram algorithm, a line chart algorithm, and a spiral chart algorithm. , stacked area chart algorithm, quantitative waveform chart algorithm, the area chart algorithm is used to display the change and development of the quantitative value within a specified time period, the bubble chart algorithm sets the variable of one of the axes to time or the data variable Animated over time, the Gantt chart algorithm is used as an organizational tool for project management, the heat map algorithm is used to display data by color changes, and the histogram algorithm is used to display data at successive intervals or at specified times the distribution of data within segments, the line chart algorithm for displaying quantitative values over successive intervals or time spans to show trends and relationships, the helix chart algorithm for plotting time-based data along an Archimedes spiral, the stacking The quantized area chart algorithm works on the same principle as the simple area chart algorithm to display multiple data series at the same time, and the quantized waveform chart algorithm is used to display changes in different categories of data over time.

Further, the KYC verification algorithm used in the process of verifying the identity of customers and customers before or during the conduct of business, the KYC verification algorithm uses AI-OCR tools to automatically capture, extract and create editable and searchable copies of customer data .

Further, the trusted identity system algorithm uses Ethereum for key distribution and management, IPFS for content addressing, ENS for human-readable information analysis, and whisper for secure data transmission channels.

Further, the consensus mechanism algorithm includes a workload proof algorithm, an equity proof algorithm, a share authorization proof mechanism algorithm, a Ripple consensus algorithm, and a practical Byzantine fault-tolerant algorithm;

The proof-of-work algorithm calculates a random number that satisfies the rules through AND-OR operation, that is, obtains the accounting right this time, sends out the data that needs to be recorded in this round, and stores it together after verification by other nodes in the entire network;

The proof-of-stake algorithm uses proof of equity to replace the proof of computing power, and the accounting right is obtained by the node with the highest equity, not the node with the highest computing power, thereby speeding up the speed of finding random numbers;

The said share authorization proof mechanism algorithm votes to select relatively reliable nodes through the proportion of assets, which can greatly reduce the number of nodes participating in verification and accounting, and can achieve consensus verification at the second level;

The Ripple consensus algorithm occurs between verification nodes. Each verification node is preconfigured with a list of trusted nodes. The nodes on the trusted node list vote for transaction completion, and each verification node will continue to receive messages from the network. After the sent transactions are verified with the local ledger data, the illegal transactions are directly discarded, and the legal transactions will be aggregated into a transaction candidate set. Each verification node sends its own transaction candidate set as a proposal to other verification nodes. After the verification node receives the proposal from other nodes, if it is not from the node on the trusted node list, it will ignore the proposal. If it is Nodes from the list of trusted nodes will compare the transaction in the proposal with the local candidate set of transactions. If there is the same transaction, the transaction will get one vote. In a certain period of time, when the transaction receives more than 50% of the votes. When there are more than 50% of the votes, the transaction will enter the next round. If there are no more than 50% of the transactions, it will be confirmed by the next consensus process. The verification node will send the transaction with more than 50% of the votes to other nodes as a proposal, and at the same time increase the threshold of the required number of votes. To 60%, repeat the calculation of proposals received by the verification node from other nodes until the threshold reaches 80%, and the verification node officially writes the transactions confirmed by the nodes in the 80% trusted node list into the local ledger data , called the last closed ledger, that is, the final state of the ledger;

The practical Byzantine fault-tolerant algorithm is a state machine replica replication algorithm, that is, a service is modeled as a state machine, and the state machine is replicated in different nodes of the distributed system. Each copy of the state machine holds the state of the service and also implements the operation of the service.

Further, the privacy protection algorithm includes a differential privacy algorithm, an Apriori algorithm, and a k-anonymity algorithm; the differential privacy algorithm is a mathematical model that quantifies the privacy loss in the data set when data is released; the Apriori algorithm is used to detect rare The original data is added with noise to form data, and then the algorithm code is added to submit it to an untrusted external platform, and the algorithm is used to eliminate the noise when it is finally returned; the k-anonymous algorithm is implemented by generalization technology and concealment technology. The generalization technique refers to a more general and abstract description of the data, making it impossible to distinguish specific values. The concealment technique refers to not publishing certain information, and by reducing the accuracy of the published data, each record is at least as good as other K in the data table. -1 records have exactly the same quasi-identifier attribute value, thereby reducing the risk of privacy leakage caused by chaining attacks.

Further, the ledger maintenance algorithm includes an endorsement stage, a sorting stage and a verification stage;

In the endorsement stage, the endorsement node checks the legitimacy of the transaction plan sent by the client, simulates the execution of the chain code to obtain the transaction result, and finally judges whether the transaction plan is supported according to the set endorsement logic; if the endorsement logic decides to support the transaction plan, The signature of the plan will be sent back to the client; by default, the endorsement logic of the endorsement node is to support the plan and sign, but the endorsement node can set the endorsement logic according to the business rules, so as to enter the team to serve the business needs. The transaction is endorsed; if the endorsement node determines that the transaction is not supported, an error message is returned to the client;

The sorting stage sorts the transactions for the sorting service and determines the time sequence relationship between the transactions; the sorting service sorts the transactions received within a period of time, and then packs the transactions after the transaction into blocks, and then divides the transactions into blocks. The block is broadcast to the members in the channel, thus ensuring the consistency of the data of all nodes;

The verification phase is to confirm that the node performs a series of verifications on the sorted transactions, including the integrity check of the transaction data, whether the transaction is repeated, whether the endorsement signature meets the requirements of the endorsement policy, and whether the transaction read-write set meets the multi-version control. All confirmation nodes verify transactions in the same order, and write legal transactions into the ledger at one time.

Further, the Docker is an advanced container engine based on LXC, the source code is hosted on Github, based on the go language and open sourced in compliance with the Apache2.0 protocol.

Further, the K8S includes a master node and a computing node; the master node includes API Server, Scheduler, Controller manager, and the API Server is the external interface of the entire system for client and other components to call, and the Scheduler is responsible for The resources inside the cluster are scheduled, and the Controller manager is responsible for managing the controller; the computing node includes the Docker, kubelet, kube-proxy, Fluentd, and Pod, and the Pod is the most basic operation unit of the K8S. Pod represents a process running in the cluster and encapsulates one or more closely related containers. The Kubelet is responsible for monitoring the Pod assigned to the computing node where the Kubelet is located, including creating, modifying, monitoring, Delete, etc., the Kube-proxy is used to provide a proxy for the Pod, and the Fluentd is used for log collection, storage and query.

Compared with the prior art, the beneficial effect of the present invention is that the blockchain-based supply chain finance AI DaaS algorithm warehouse platform provided by the present invention includes an application layer, an open interface layer, an algorithm warehouse service layer, and a data layer that are connected in sequence. As well as the infrastructure layer, the application layer is used to provide supply chain financial services based on blockchain, the open interface layer is used to provide the application layer with services to connect the algorithm warehouse service layer, and the data layer is used to provide data service support for the algorithm warehouse service layer. , the infrastructure layer is used to provide development support for the data layer; the invention solves the problem of algorithm fusion and sharing of the supply chain finance of the blockchain. .

Description of drawings

In order to illustrate the technical solutions in the embodiments of the present invention more clearly, the following briefly introduces the drawings required in the embodiments. Obviously, the drawings in the following description are only some embodiments of the present invention. For those of ordinary skill in the art, other drawings can also be obtained from these drawings without any creative effort.

1 is a system structure diagram of a blockchain-based supply chain finance AI DaaS algorithm warehouse platform provided by an embodiment of the present invention.

Detailed ways

In order to make the objectives, technical solutions and advantages of the present invention clearer, the present invention will be further described in detail below with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the present invention, but not to limit the present invention.

The same or similar numbers in the drawings of this embodiment correspond to the same or similar components; in the description of the present invention, it should be understood that if there are terms such as "upper", "lower", "left", "right", etc. The indicated orientation or positional relationship is based on the orientation or positional relationship shown in the accompanying drawings, which is only for the convenience of describing the present invention and simplifying the description, rather than indicating or implying that the indicated device or element must have a specific orientation or a specific orientation. structure and operation, so the terms describing the positional relationship in the accompanying drawings are only used for exemplary illustration, and should not be construed as a limitation on this patent, and those of ordinary skill in the art can understand the specific meanings of the above terms according to specific situations.

The technical solutions of the present invention will be described in detail below with reference to the accompanying drawings and specific embodiments.

Referring to Figure 1, the blockchain-based supply chain finance AI DaaS algorithm warehouse platform provided by the present invention includes an application layer, an open interface layer, an algorithm warehouse service layer, a data layer and an infrastructure layer that are sequentially connected. In order to provide supply chain financial services based on blockchain, the open interface layer is used to provide the application layer with the service of connecting the algorithm warehouse service layer, and the data layer is used to provide data for the algorithm warehouse service layer. service support, the infrastructure layer is used to provide development support for the data layer;

The blockchain-based supply chain finance AI DaaS algorithm warehouse platform provided by the above technical solution includes an application layer, an open interface layer, an algorithm warehouse service layer, a data layer and an infrastructure layer that are connected in sequence. The supply chain financial service of the chain, the open interface layer is used to provide the application layer with the service of connecting the algorithm warehouse service layer, the data layer is used to provide data service support for the algorithm warehouse service layer, and the infrastructure layer is used to provide development support for the data layer; The invention solves the problem of algorithm fusion and sharing of supply chain finance of block chain.

As an embodiment of the present invention, the data visualization algorithm includes a parallel visualization algorithm, an in-situ visualization algorithm, and a time series data visualization algorithm;

As an embodiment of the present invention, the KYC verification algorithm is used in the process of verifying customers and customer identities before or during the conduct of business, and the KYC verification algorithm uses AI-OCR tools to automatically capture, extract and create editable and Searchable copy of customer data.

As an embodiment of the present invention, the trusted identity system algorithm uses Ethereum for key distribution and management, IPFS for content addressing, ENS for human-readable information parsing, and whisper as a secure data transmission channel.

As an embodiment of the present invention, the consensus mechanism algorithm includes a workload proof algorithm, a rights proof algorithm, a share authorization proof mechanism algorithm, a Ripple consensus algorithm, and a practical Byzantine fault-tolerant algorithm;

As an embodiment of the present invention, the privacy protection algorithm includes a differential privacy algorithm, an Apriori algorithm, and a k-anonymity algorithm; the differential privacy algorithm is a mathematical model that quantifies the privacy loss in a data set when data is released; the The Apriori algorithm is used to detect rare data classes, add noise to the original data to form data, then add the algorithm code and submit it to an untrusted external platform, and use the algorithm to eliminate the noise when it finally returns; the k-anonymous algorithm uses generalization technology and It is realized by concealment technology. The generalization technology refers to a more general and abstract description of the data, making it impossible to distinguish specific values. The concealment technology means that certain information is not released, and by reducing the accuracy of the released data, each record is at least as good as the one. The other K-1 records in the data table have exactly the same quasi-identifier attribute value, thereby reducing the risk of privacy leakage caused by chaining attacks.

As an embodiment of the present invention, the ledger maintenance algorithm includes an endorsement phase, a sorting phase, and a verification phase;

Specifically, the Docker is an advanced container engine based on LXC, the source code is hosted on Github, based on the go language and open sourced in compliance with the Apache2.0 protocol.

Specifically, the K8S includes a master node and a computing node; the master node includes an API Server, a Scheduler, and a Controller manager, and the API Server is the external interface of the entire system, which is called by clients and other components, and the Scheduler is responsible for The resources inside the cluster are scheduled, and the Controller manager is responsible for managing the controller; the computing node includes the Docker, kubelet, kube-proxy, Fluentd, and Pod, and the Pod is the most basic operation unit of the K8S. Pod represents a process running in the cluster and encapsulates one or more closely related containers. The Kubelet is responsible for monitoring the Pod assigned to the computing node where the Kubelet is located, including creating, modifying, monitoring, Delete, etc., the Kube-proxy is used to provide a proxy for the Pod, and the Fluentd is used for log collection, storage and query.

Preferably, the functions and business processes of all modules involved in the present invention adopt existing and public functions and business processes; the architectures of all modules involved in the present invention adopt existing and public architectures. Architecture mode; the implementation modes of all modules involved in the present invention adopt open, mature, open source program architecture and program codes, and those skilled in the art can easily adopt existing, open source, and open source programs according to the functions described in this technical solution. The program architecture and program code implementation.

The embodiments of the present invention have been described in detail above, but the creation of the present invention is not limited to the present embodiment. Those skilled in the art can also make many equivalent modifications or substitutions without departing from the spirit of the present invention. These equivalent modifications or substitutions are all included within the protection scope defined by the claims of the present application.

Claims

The blockchain-based supply chain finance AI DaaS algorithm warehouse platform is characterized in that it includes an application layer, an open interface layer, an algorithm warehouse service layer, a data layer and an infrastructure layer that are connected in sequence. The supply chain financial service of the blockchain, the open interface layer is used to provide the application layer with the service of connecting the algorithm warehouse service layer, the data layer is used to provide data service support for the algorithm warehouse service layer, so The infrastructure layer is used to provide development support for the data layer;

The application layer includes a letter of credit module, a non-recourse financing module, a supply chain finance module, a smart retail module, a link management module, a smart contract script module, a smart contract template module, and a smart contract management platform. It is used to provide a letter of credit business function, the non-recourse financing module is used to provide a non-recourse financing business function, the supply chain financial module is used to provide support services for trusted credit transfer, and the smart retail module is used for In order to provide smart retail business functions, the link management module is used to provide data link management functions, the smart contract script module is used to provide the automatic writing function of smart contract scripts, and the smart contract template module is used to provide intelligent The function of the contract, the smart contract management platform is used to provide the management function of the smart contract;

The open interface layer includes a chain code protocol interface, a blockchain API service interface, an identity authentication service interface, and a light client interface. The chain code protocol interface is a blockchain access interface, and the blockchain API service interface uses In order to provide all the functions related to the blockchain so that the application system only needs to call the relevant interface to become a blockchain application, the identity authentication service interface is used to provide the function of verifying user information, and the light client interface is used to Provide users with a secure and decentralized way to access and interact with the blockchain without synchronizing the entire blockchain;

The algorithm warehouse service layer includes a data visualization algorithm module, a KYC verification algorithm module, a trusted identity system algorithm module, a consensus mechanism algorithm module, a privacy protection algorithm module, and a ledger maintenance algorithm module, and the data visualization algorithm module is used for data visualization algorithms. The KYC verification algorithm module is used for the storage and application of the KYC verification algorithm, the trusted identity system algorithm module is used for the storage and application of the trusted identity system algorithm, and the consensus mechanism algorithm module is used for consensus The storage and application of the mechanism algorithm, the privacy protection algorithm module is used for the storage and application of the privacy protection algorithm, and the account book maintenance algorithm module is used for the storage and application of the account book maintenance algorithm;

The data layer includes a block data sharing module, a block data governance module, and a block data development module. The block data sharing module is used to provide data sharing functions for different modules, and the block data governance module is used to provide different modules. Data governance function, the block data development module is used to provide data development function for different modules;

The infrastructure layer includes Docker, Docker-compose, K8S, the Docker is an open source application container engine, the Docker-compose is an application tool for defining and running multi-container Docker, and the K8S is a container-based cluster management The platform is used to manage Docker and containers.
The blockchain-based supply chain finance AI DaaS algorithm warehouse platform according to claim 1, wherein the data visualization algorithm comprises a parallel visualization algorithm, an in-situ visualization algorithm and a time series data visualization algorithm;

The parallel visualization algorithm includes a task parallel algorithm, a pipeline parallel algorithm, and a data parallel algorithm; the task parallel algorithm divides the visualization process into independent subtasks, and there is no data dependency between the running subtasks; the pipeline The parallel algorithm uses streaming to read data fragments, divides the visualization process into multiple stages, and executes each stage in parallel by the computer to speed up the processing process; the data parallel algorithm is a single-program multiple-data method, which divides the data into multiple subsets, and then use the subset as the granularity to execute the program in parallel to process the different subsets;

The in-situ visualization algorithm is to generate visualization in the process of numerical simulation and is used to alleviate the output bottleneck of large-scale numerical simulation, including image in-situ visualization algorithm, distributed data in-situ visualization algorithm, compressed data in-situ visualization algorithm and feature in-situ visualization. The in-situ visualization algorithm of the image is to map the data into visualization during the numerical simulation process and save it as an image; the in-situ visualization algorithm of the distribution data is calculated according to the statistical index defined by the user during the numerical simulation Statistical indicators are saved, followed by statistical data visualization; the in-situ visualization of the compressed data adopts a compression algorithm to reduce the output scale of the numerical simulation data, and the compressed data is used as the input of the subsequent visualization processing; the feature in-situ visualization algorithm is in the numerical simulation process. Extract features and save them, and use feature data as input for subsequent visualization processing;

The time series data visualization algorithm is used to establish a prediction model, perform predictive analysis and user behavior analysis, including an area chart algorithm, a bubble chart algorithm, a Gantt chart algorithm, a heat map algorithm, a histogram algorithm, a line chart algorithm, and a spiral chart algorithm. , stacked area chart algorithm, quantitative waveform chart algorithm, the area chart algorithm is used to display the change and development of the quantitative value within a specified time period, the bubble chart algorithm sets the variable of one of the axes to time or the data variable Animated over time, the Gantt chart algorithm is used as an organizational tool for project management, the heat map algorithm is used to display data by color changes, and the histogram algorithm is used to display data at successive intervals or at specified times the distribution of data within segments, the line chart algorithm for displaying quantitative values over successive intervals or time spans to show trends and relationships, the helix chart algorithm for plotting time-based data along an Archimedes spiral, the stacking The quantized area chart algorithm works on the same principle as the simple area chart algorithm to display multiple data series at the same time, and the quantized waveform chart algorithm is used to display changes in different categories of data over time.
The blockchain-based supply chain finance AI DaaS algorithm warehouse platform according to claim 1, wherein the KYC verification algorithm is used for the process of verifying the identity of customers and customers before or during the conduct of business, and the KYC verification algorithm Validation algorithms employ AI-OCR tools to automatically capture, extract and create editable and searchable copies of customer data.
The blockchain-based supply chain finance AI DaaS algorithm warehouse platform according to claim 1, wherein the trusted identity system algorithm adopts Ethereum for key distribution and management, IPFS is responsible for content addressing, and ENS is responsible for For human-readable information parsing, whisper acts as a secure data transmission channel.
The blockchain-based supply chain finance AI DaaS algorithm warehouse platform according to claim 1, wherein the consensus mechanism algorithm includes a workload proof algorithm, a rights proof algorithm, a share authorization proof mechanism algorithm, a Ripple consensus algorithm, Practical Byzantine fault-tolerant algorithm;

The proof-of-work algorithm calculates a random number that satisfies the rules through AND-OR operation, that is, obtains the accounting right this time, sends out the data that needs to be recorded in this round, and stores it together after verification by other nodes in the entire network;

The proof-of-stake algorithm uses proof of equity to replace the proof of computing power, and the accounting right is obtained by the node with the highest equity, not the node with the highest computing power, thereby speeding up the speed of finding random numbers;

The said share authorization proof mechanism algorithm votes to select relatively reliable nodes through the proportion of assets, which can greatly reduce the number of nodes participating in verification and accounting, and can achieve consensus verification at the second level;

The Ripple consensus algorithm occurs between verification nodes. Each verification node is preconfigured with a list of trusted nodes. The nodes on the trusted node list vote for transaction completion, and each verification node will continue to receive messages from the network. After the sent transactions are verified with the local ledger data, the illegal transactions are directly discarded, and the legal transactions will be aggregated into a transaction candidate set. Each verification node sends its own transaction candidate set as a proposal to other verification nodes. After the verification node receives the proposal from other nodes, if it is not from the node on the trusted node list, it will ignore the proposal. If it is Nodes from the list of trusted nodes will compare the transaction in the proposal with the local candidate set of transactions. If there is the same transaction, the transaction will get one vote. In a certain period of time, when the transaction receives more than 50% of the votes. When there are more than 50% of the votes, the transaction will enter the next round. If there are no more than 50% of the transactions, it will be confirmed by the next consensus process. The verification node will send the transaction with more than 50% of the votes to other nodes as a proposal, and at the same time increase the threshold of the required number of votes. To 60%, repeat the calculation of proposals received by the verification node from other nodes until the threshold reaches 80%, and the verification node officially writes the transactions confirmed by the nodes in the 80% trusted node list into the local ledger data , called the last closed ledger, that is, the final state of the ledger;

The practical Byzantine fault-tolerant algorithm is a state machine replica replication algorithm, that is, a service is modeled as a state machine, and the state machine is replicated in different nodes of the distributed system. Each copy of the state machine holds the state of the service and also implements the operation of the service.
The blockchain-based supply chain finance AI DaaS algorithm warehouse platform according to claim 1, wherein the privacy protection algorithm includes a differential privacy algorithm, an Apriori algorithm, and a k-anonymity algorithm; the differential privacy algorithm is a pair of Mathematical model for quantifying the privacy loss in the data set when the data is released; the Apriori algorithm is used to detect rare data classes, add noise to the original data to form data, and then add the algorithm code and submit it to an untrusted external platform, and when the final return Then use an algorithm to eliminate noise; the k-anonymous algorithm is realized by generalization technology and concealment technology. The generalization technology refers to a more general and abstract description of the data, so that the specific values cannot be distinguished. For some information, by reducing the precision of the published data, each record has at least the same quasi-identifier attribute value as other K-1 records in the data table, thereby reducing the risk of privacy leakage caused by link attacks.
The blockchain-based supply chain finance AI DaaS algorithm warehouse platform according to claim 1, wherein the ledger maintenance algorithm includes an endorsement stage, a sorting stage and a verification stage;

In the endorsement stage, the endorsement node checks the legitimacy of the transaction plan sent by the client, simulates the execution of the chain code to obtain the transaction result, and finally judges whether the transaction plan is supported according to the set endorsement logic; if the endorsement logic decides to support the transaction plan, The signature of the plan will be sent back to the client; by default, the endorsement logic of the endorsement node is to support the plan and sign, but the endorsement node can set the endorsement logic according to the business rules, so as to enter the team to serve the business needs. The transaction is endorsed; if the endorsement node determines that the transaction is not supported, an error message is returned to the client;

The sorting stage sorts the transactions for the sorting service and determines the time sequence relationship between the transactions; the sorting service sorts the transactions received within a period of time, and then packs the transactions after the transaction into blocks, and then divides the transactions into blocks. The block is broadcast to the members in the channel, thus ensuring the consistency of the data of all nodes;

The verification phase is to confirm that the node performs a series of verifications on the sorted transactions, including the integrity check of the transaction data, whether the transaction is repeated, whether the endorsement signature meets the requirements of the endorsement policy, and whether the transaction read-write set meets the multi-version control. All confirmation nodes verify transactions in the same order, and write legal transactions into the ledger at one time.
The blockchain-based supply chain finance AI DaaS algorithm warehouse platform according to claim 1, wherein the Docker is an LXC-based advanced container engine, and the source code is hosted on Github, based on go language and complies with Apache2. 0 protocol open source.
The blockchain-based supply chain finance AI DaaS algorithm warehouse platform according to claim 1, wherein the K8S includes a master node and a computing node; the master node includes an API Server, a Scheduler, and a Controller manager, and the API Server is the external interface of the entire system, which is called by clients and other components. The Scheduler is responsible for scheduling the resources inside the cluster, and the Controller manager is responsible for managing the controller; the computing nodes include the Docker, kubelet, kube -proxy, Fluentd, Pod. The Pod is the most basic operation unit of the K8S. The Pod represents a process running in the cluster and encapsulates one or more closely related containers. The Kubelet is responsible for monitoring assignments To the Pod on the computing node where the Kubelet is located, including creation, modification, monitoring, deletion, etc., the Kube-proxy is used to provide a proxy for the Pod, and the Fluentd is used for log collection, storage and query .