A product information tamper-proof management method and system

Abstract

The product information anti-tampering management method and system provided by the present invention relate to the technical field of product information anti-tampering management, and solve the problem of traceability failure and easy forgery caused by the misuse of the correct product number. The method includes: constructing a blockchain for comprehensive management of product batch information; the accounting node in the blockchain records the product information of each product; the traceability query and AI verification server uses the quality inspection data of all products under the same production batch to train the AI ​​model, and obtains the batch fingerprint data of the corresponding production batch and its matching AI model; when the customer needs to trace the product, the product traceability client retrieves all product information and the corresponding batch fingerprint data and AI model corresponding to the product number of the product to be traced from the traceability query and AI verification server on the blockchain, and obtains the traceability result based on the obtained product information, the corresponding batch fingerprint data, the AI ​​model and the information of the product to be traced.

Description

Technical Field

[0001] The present invention relates to the technical field of product information tamper-proof management, and in particular to a product information tamper-proof management method and system. Background Art

[0002] At present, batch management can be achieved through the MES system, which can automatically generate production batch numbers and record and query each production batch. During the production process, the MES system will automatically record the production status of each batch. If a problem occurs, the MES system can quickly locate the problem batch through the traceability function, so as to deal with the problem in time and ensure the quality and safety of the product.

[0003] However, in actual application, if the records in the batch management server are not tampered with, but someone uses the correct product number, the traceability process will fail. In addition, since the traceability management client may be a mobile device, it cannot perform more complex anti-counterfeiting functions. Summary of the invention

[0004] In view of the above analysis, an embodiment of the present invention aims to provide a product information tamper-proof management method to solve the problem that the existing batch management server fails to trace the product and is easy to be forged due to someone misusing the correct product number during use.

[0005] In one aspect, the present invention discloses a product information tamper-proof management method, the method comprising:

[0006] Build a blockchain for comprehensive management of product batch information, and add MES servers, DNC servers, CNC machine tools, quality inspection servers, traceability query and AI verification servers to the blockchain;

[0007] The accounting nodes in the blockchain interact with the MES server, DNC server, CNC machine tool and quality inspection server respectively to record the product information of each product, including: production batch number, product number, production program version number, CNC machine tool number and quality inspection data; each production batch uniquely corresponds to a production batch number;

[0008] The traceability query and AI verification server obtains the product information of all products under the same production batch from the blockchain, and uses the quality inspection data of all products under the same production batch to train the AI ​​model, and obtains the batch fingerprint data of the corresponding production batch and its matching AI model;

[0009] When a customer needs to trace a product, the product traceability client is added to the blockchain. The product traceability client retrieves all product information and corresponding batch fingerprint data and AI models corresponding to the product number of the product to be traced from the traceability query and AI verification server on the blockchain, and obtains the traceability result based on the acquired product information, corresponding batch fingerprint data, AI model and information of the product to be traced.

[0010] On the basis of the above scheme, the present invention also makes the following improvements:

[0011] Furthermore, based on the acquired product information, the corresponding batch fingerprint data, the AI ​​model and the information of the product to be traced, the traceability result is obtained and the following is executed:

[0012] Obtain product information of the product to be traced according to the nameplate of the product to be traced;

[0013] The obtained product information is compared with the product information of the product to be traced one by one. If they are not all consistent, the traceability of the product to be traced fails;

[0014] If all correspondences are consistent, the following is also executed: using the retrieved AI model to process the quality inspection data of the product to be traced to obtain the batch fingerprint characteristics of the product to be traced; if the batch fingerprint characteristics of the product to be traced match the retrieved batch fingerprint characteristics, the traceability of the product to be traced is successful; otherwise, the traceability of the product to be traced fails.

[0015] Furthermore, in the traceability query and AI verification server, the AI ​​model is trained in the following ways:

[0016] Perform feature extraction on all quality inspection data of the same production batch to obtain feature vectors of corresponding quality inspection data;

[0017] Use the feature vectors of all quality inspection data of the same production batch as positive sample data to train the AI ​​model to learn the common features in the positive sample data;

[0018] After the AI ​​model training is completed, the common features in the positive sample data extracted by the AI ​​model are used as the batch fingerprint data of the corresponding production batch, and the trained AI model is used as the AI ​​model that matches the batch fingerprint data of the corresponding production batch.

[0019] Furthermore, the feature vector of the quality inspection data is obtained by performing the following operations:

[0020] Arrange multiple dimensions involved in the quality inspection standards of products of the same production batch in a fixed manner to form a feature extraction template;

[0021] Feature extraction is performed on each quality inspection data in the same production batch according to the feature extraction template to obtain the feature vector of the corresponding quality inspection data.

[0022] Furthermore, the accounting nodes in the blockchain interact with the MES server to monitor the production process of the product and perform:

[0023] The MES server broadcasts the production schedule to the blockchain; each production task in the production schedule includes: a production batch number, and a preset product number of the product to be processed;

[0024] The accounting node receives the records published by the MES server and determines whether the production batch number and product number of each production task are recorded in the historical block. If not, the corresponding production task is marked as the first record and written into the blockchain.

[0025] Furthermore, the accounting nodes in the blockchain interact with the DNC server to supervise the production process of the product and perform:

[0026] The DNC server receives the production task from the blockchain, successfully sends the production program of the corresponding production task to the CNC machine tool, and broadcasts the corresponding production task and its production program version number to the blockchain;

[0027] The accounting node receives the record published by the DNC server. If the accounting node finds that the corresponding production task is recorded in the historical block but the production program version number of the corresponding production task is not recorded, the corresponding production task and its production program version number published by the first DNC ​​server broadcast in the blockchain are marked as the second record and written into the blockchain.

[0028] Furthermore, the accounting nodes in the blockchain interact with CNC machine tools to monitor the production process of products and perform:

[0029] The data machine tool processes the product based on the production program and assigns a preset product number to the processed product; then, the production batch number in the corresponding production task, the product number of the processed product, and the data machine tool number of the corresponding production task are broadcast to the blockchain;

[0030] The accounting node receives the records released by the CNC machine tool, and determines whether the first record and the second record in the history block include the production batch number released by the CNC machine tool and the product number of the processed product. If so, the production batch number in the corresponding production task, the product number of the processed product, and the data machine tool number of the corresponding production task are marked as the third record and written into the blockchain.

[0031] Furthermore, the accounting nodes in the blockchain interact with the quality inspection server to monitor the production process of the product and perform:

[0032] The quality inspection server conducts quality inspection on the processed products and obtains the quality inspection data; and broadcasts the production batch number, the product number of the processed product and its quality inspection data to the blockchain;

[0033] The accounting node receives the records released by the quality inspection server, and determines whether the first record, the second record, and the third record in the history block include the production batch number and the product number of the processed product released by the quality inspection server. If so, the production batch number in the corresponding production task, the product number of the processed product and its quality inspection data are marked as the fourth record and written into the blockchain.

[0034] On the other hand, the present invention discloses a product information tamper-proof management system, the system comprising:

[0035] The blockchain for comprehensive management of product batch information is used to record the product information of each product, including: production batch number, product number, production program version number, CNC machine tool number and quality inspection data; each production batch uniquely corresponds to a production batch number;

[0036] The traceability query and AI verification server is used to join the blockchain and obtain the product information of all products under the same production batch from the blockchain, and use the quality inspection data of all products under the same production batch to train the AI ​​model to obtain the batch fingerprint data of the corresponding production batch and its matching AI model;

[0037] The product traceability client is used to join the blockchain when the customer needs to trace the product. The product traceability client retrieves all product information and corresponding batch fingerprint data and AI model corresponding to the product number of the product to be traced from the traceability query and AI verification server on the blockchain, and obtains the traceability result based on the acquired product information, corresponding batch fingerprint data, AI model and information of the product to be traced.

[0038] On the basis of the above scheme, the present invention also makes the following improvements:

[0039] Furthermore, the system also includes an MES server, a DNC server, a CNC machine tool and a quality inspection server;

[0040] The MES server, DNC server, CNC machine tool and quality inspection server are added to the blockchain, and interact with the MES server, DNC server, CNC machine tool and quality inspection server respectively through the accounting nodes in the blockchain to supervise the production process of the products and record the production batch number, product number, production program version number, CNC machine tool number and quality inspection data of each product.

[0041] Compared with the prior art, the present invention can achieve at least one of the following beneficial effects:

[0042] The product information tamper-proof management method provided by the present invention records the production batch number, product number, production program version number, CNC machine number and quality inspection data of each product on the blockchain, and uses the quality inspection data of all products under the same production batch to train the AI ​​model, obtain the batch fingerprint data of the corresponding production batch and its matching AI model, combine the production batch number, product number, production program version number, CNC machine number, quality inspection data, batch fingerprint data and its matching AI model of each product, form all product data of the corresponding product, and provide conditions for product tracing. When product tracing is required, the information of the product to be traced is compared with the obtained product data, especially the AI ​​model and batch fingerprint features are involved in the tracing judgment, so that a more accurate tracing result can be obtained, which can solve the problem that the existing batch management server fails to trace the source and is easy to be forged due to the misuse of the correct product number during use.

[0043] The corresponding product information anti-tampering management system is implemented based on the same technical concept as the above method and has corresponding technical effects.

[0044] In the present invention, the above-mentioned technical solutions can also be combined with each other to achieve more preferred combination solutions. Other features and advantages of the present invention will be described in the subsequent description, and some advantages can become obvious from the description, or can be understood by practicing the present invention. The purpose and other advantages of the present invention can be realized and obtained through the contents particularly pointed out in the description and the drawings. BRIEF DESCRIPTION OF THE DRAWINGS

[0045] The accompanying drawings are only used for the purpose of illustrating specific embodiments and are not to be considered as limiting the present invention. In the entire drawings, the same reference symbols represent the same components;

[0046] Figure 1 A flowchart of a product information tamper-proof management method provided in Example 1 of the present invention;

[0047] Figure 2 This is a schematic diagram of the structure of the product information tamper-proof management system provided in Example 2 of the present invention. DETAILED DESCRIPTION

[0048] The preferred embodiments of the present invention will be described in detail below in conjunction with the accompanying drawings, wherein the accompanying drawings constitute a part of this application and are used together with the embodiments of the present invention to illustrate the principles of the present invention, but are not used to limit the scope of the present invention.

[0049] In this embodiment, the blockchain is introduced into the comprehensive management of product batch information of the MES system. From production scheduling, DNC server to issuing production programs, to CNC machine tools starting production, to final product quality inspection and acceptance, the process data generated in the entire process of product production are uploaded to the blockchain for storage. The product traceability client is also added to the blockchain. When receiving the traceability query, it obtains relevant data from the blockchain for traceability verification, and presents the traceability verification results to the inquiring party. In order to prevent someone from using false products to mark the correct product number, this embodiment sets up a traceability query and AI verification server, performs AI modeling on the quality inspection data of all products in the same production batch, and extracts the fingerprint feature data of all products in the production batch, so as to compare the fingerprint feature data of the product to be traced with the fingerprint feature data of the corresponding production batch products, and conduct counterfeit tracing.

[0050] A specific embodiment 1 of the present invention discloses a product information tamper-proof management method, the flow chart is as follows: Figure 1 As shown, the implementation steps of the method are specifically described as follows.

[0051] Step S1: Build a blockchain for comprehensive management of product batch information, and add MES servers, DNC servers, CNC machine tools, quality inspection servers, and traceability query and AI verification servers to the blockchain;

[0052] Step S2: The accounting nodes in the blockchain interact with the MES server, DNC server, CNC machine tool and quality inspection server respectively to record the product information of each product, including: production batch number, product number, production program version number, CNC machine tool number and quality inspection data; each production batch uniquely corresponds to a production batch number;

[0053] Step S3: The traceability query and AI verification server obtains the product information of all products under the same production batch from the blockchain, and uses the quality inspection data of all products under the same production batch to train the AI ​​model, and obtains the batch fingerprint data of the corresponding production batch and its matching AI model;

[0054] Step S4: When the customer needs to trace the product, the product traceability client is added to the blockchain. The product traceability client retrieves all product information corresponding to the product number of the product to be traced and the corresponding batch fingerprint data and AI model from the traceability query and AI verification server on the blockchain, and obtains the traceability result based on the obtained product information, the corresponding batch fingerprint data, the AI ​​model and the information of the product to be traced.

[0055] In this embodiment, the blockchain monitors the production process of the product by interacting with the MES server, DNC server, CNC machine tools and quality inspection server. The specific process is described as follows.

[0056] First, the accounting node in the blockchain interacts with the MES server to monitor the production process of the product and execute:

[0057] Step S21: The MES server broadcasts the production schedule to the blockchain; each production task in the production schedule includes: a production batch number, and a preset product number of the product to be processed. It should be noted that in this embodiment, each production task uniquely determines a production batch number; in each production task, each product to be processed uniquely determines a product number; the combination of the production batch number and the product number can uniquely determine a product to be processed.

[0058] Step S22: The accounting node receives the records published by the MES server, and determines whether the production batch number and product number of each production task are recorded in the historical block. If not, the corresponding production task is marked as the first record and written into the blockchain.

[0059] In the specific implementation process, the accounting node needs to ensure the global uniqueness of the production batch number and product number. If the production batch number and product number already exist in the blockchain, the corresponding production task will not be written into the blockchain.

[0060] After the production task is written into the blockchain, the DNC server can select the appropriate production task from the blockchain according to its own situation. At this time, the accounting node in the blockchain interacts with the DNC server to supervise the production process of the product and execute:

[0061] Step S23: The DNC server receives the production task from the blockchain, successfully sends the production program of the corresponding production task to the CNC machine tool, and broadcasts the corresponding production task and its production program version number to the blockchain;

[0062] Step S24: The accounting node receives the record published by the DNC server. If the accounting node finds that the corresponding production task is recorded in the historical block, but the production program version number of the corresponding production task is not recorded, the corresponding production task and its production program version number published by the first DNC ​​server broadcasted in the blockchain are marked as the second record and written into the blockchain. Otherwise, the record published by the DNC server is invalid and directly ignored.

[0063] After the DNC server successfully writes the corresponding production task and its production program version number into the blockchain, the CNC machine tool can start processing the product. In this process, the accounting node in the blockchain interacts with the CNC machine tool to monitor the production process of the product and execute:

[0064] Step S25: The data machine tool processes the product based on the production program and assigns a preset product number to the processed product; then, the production batch number in the corresponding production task, the product number of the processed product, and the data machine tool number of the corresponding production task are broadcast to the blockchain;

[0065] Step S26: The accounting node receives the record released by the CNC machine tool, and determines whether the first record and the second record in the history block include the production batch number released by the CNC machine tool and the product number of the processed product. If so, the production batch number in the corresponding production task, the product number of the processed product, and the data machine tool number of the corresponding production task are marked as the third record and written into the blockchain; otherwise, the record released by the CNC machine tool is invalid and is directly ignored.

[0066] After the product is processed, it needs to be inspected for quality. At this time, the accounting node in the blockchain interacts with the quality inspection server to supervise the production process of the product and execute:

[0067] Step S27: The quality inspection server performs quality inspection on the processed product to obtain quality inspection data; and broadcasts the production batch number, the product number of the processed product and its quality inspection data to the blockchain;

[0068] Step S28: The accounting node receives the record published by the quality inspection server, and determines whether the first record, the second record, and the third record in the history block include the production batch number and the product number of the processed product published by the quality inspection server. If so, the production batch number in the corresponding production task, the product number of the processed product and its quality inspection data are marked as the fourth record and written into the blockchain. Otherwise, the record published by the quality inspection server is invalid and is directly ignored.

[0069] In step S3, in the traceability query and AI verification server, the AI ​​model is trained in the following way:

[0070] Step S31: extracting features from all quality inspection data of the same production batch to obtain feature vectors of corresponding quality inspection data;

[0071] Step S32: using the feature vectors of all quality inspection data of the same production batch as positive sample data, training the AI ​​model to learn the common features in the positive sample data;

[0072] Step S33: After the AI ​​model training is completed, the common features in the positive sample data extracted by the AI ​​model are used as the batch fingerprint data of the corresponding production batch, and the trained AI model is used as the AI ​​model that matches the batch fingerprint data of the corresponding production batch.

[0073] Preferably, in step S31, the feature vector of the quality inspection data is obtained by performing the following operations: sorting the multiple dimensions involved in the quality inspection standards of products of the same production batch in a fixed manner to form a feature extraction template; performing feature extraction on each quality inspection data in the same production batch according to the feature extraction template to obtain the feature vector of the corresponding quality inspection data.

[0074] In the specific implementation process, we can select a suitable AI model, such as a deep learning model (using DNN, CNN, Transformer, etc.) based on actual experience or by conducting a large number of experiments. We can use the feature vectors of all quality inspection data of the same batch as positive sample data to train the AI ​​model to learn the common features in the positive sample data. These common features can characterize the common properties of all products produced in the same production batch. In addition, unsupervised learning methods such as autoencoders can be used to automatically learn useful feature representations from the feature vectors of all quality inspection data of the same production batch. In the process of training the AI ​​model, we can use appropriate evaluation indicators, such as accuracy, recall rate, F1 score, etc., to evaluate the performance of the AI ​​model to ensure that the AI ​​model can effectively extract the common features of the positive sample data and has good generalization ability. And according to the evaluation results, the AI ​​model is adjusted and optimized to improve its accuracy and practicality.

[0075] In practical applications, both the feature extraction and model training of AI models need to be combined with specific production requirements and data characteristics to select appropriate model architectures and training strategies. At the same time, the evaluation and optimization of AI models is also a continuous process that needs to be constantly adjusted based on training feedback. In this way, AI models can help companies to conduct quality control more effectively and increase the difficulty of product counterfeiting. The AI ​​model that matches the batch fingerprint feature of the current production batch can make the quality inspection data of all products in the current production batch conform to the batch fingerprint feature of the current production batch (that is, the batch fingerprint features obtained by inputting the feature vectors of each quality inspection data of the same production batch into the corresponding AI model respectively match the batch fingerprint features of the current production batch), while the quality inspection data of products from other production batches do not conform to the batch fingerprint features of the current production batch (that is, the batch fingerprint features obtained by inputting the feature vectors of the quality inspection data of other production batches into the corresponding AI model do not match the batch fingerprint features of the current production batch). This makes it possible to use the trained AI model and batch fingerprint features of the corresponding batch to trace the product.

[0076] In step S4, based on the acquired product information, the corresponding batch fingerprint data, the AI ​​model and the information of the product to be traced, the traceability result is obtained and the following is executed:

[0077] Step S41: obtaining product information of the product to be traced according to the nameplate of the product to be traced;

[0078] Step S42: Compare the acquired product information with the product information of the product to be traced one by one.

[0079] If not all correspond, the traceability of the product to be traced fails;

[0080] If all correspondences are consistent, the following is also executed: using the retrieved AI model to process the quality inspection data of the product to be traced to obtain the batch fingerprint characteristics of the product to be traced; if the batch fingerprint characteristics of the product to be traced match the retrieved batch fingerprint characteristics, the traceability of the product to be traced is successful; otherwise, the traceability of the product to be traced fails.

[0081] In a specific embodiment 2 of the present invention, a product information tamper-proof management system is also disclosed, and the structural diagram is shown in FIG. Figure 2 As shown, the system includes:

[0082] Blockchain for comprehensive management of product batch information (i.e. Figure 2 The "Product Batch and Traceability Management Blockchain" in the 2018 China International Import Expo is used to record the product information of each product, including: production batch number, product number, production program version number, CNC machine tool number and quality inspection data; each production batch has a unique production batch number;

[0083] The traceability query and AI verification server is used to join the blockchain and obtain the product information of all products under the same production batch from the blockchain, and use the quality inspection data of all products under the same production batch to train the AI ​​model to obtain the batch fingerprint data of the corresponding production batch and its matching AI model;

[0084] The product traceability client is used to join the blockchain when the customer needs to trace the product. The product traceability client retrieves all product information and corresponding batch fingerprint data and AI model corresponding to the product number of the product to be traced from the traceability query and AI verification server on the blockchain, and obtains the traceability result based on the acquired product information, corresponding batch fingerprint data, AI model and information of the product to be traced.

[0085] During the specific implementation process, the product traceability client publishes a query record of the product number of the product to be traced on the blockchain. Based on the query record, the accounting node in the blockchain retrieves all product data corresponding to the product number of the product to be traced from the traceability query and AI verification server on the blockchain, and broadcasts it in the blockchain. The product traceability client downloads all product data corresponding to the product number of the product to be traced from the blockchain.

[0086] Preferably, the system also includes an MES server, a DNC server, a CNC machine tool (i.e., Figure 2The MES server, DNC server, CNC machine tool and quality inspection server are added to the blockchain, and interact with the MES server, DNC server, CNC machine tool and quality inspection server through the accounting nodes in the blockchain to supervise the production process of the products and record the production batch number, product number, production program version number, CNC machine tool number and quality inspection data of each product.

[0087] The specific implementation process of the embodiment of the present invention can be referred to the above method embodiment, and this embodiment will not be repeated here. Since the principle of this embodiment is the same as that of the above method embodiment, the system also has the corresponding technical effects of the above method embodiment.

[0088] Those skilled in the art will appreciate that all or part of the processes of the above-mentioned embodiments can be implemented by instructing related hardware through a computer program, and the program can be stored in a computer-readable storage medium, wherein the computer-readable storage medium is a disk, an optical disk, a read-only storage memory, or a random access memory, etc.

[0089] The above description is only a preferred specific implementation manner of the present invention, but the protection scope of the present invention is not limited thereto. Any changes or substitutions that can be easily conceived by any technician familiar with the technical field within the technical scope disclosed by the present invention should be covered within the protection scope of the present invention.

Claims

1. A product information tamper-proof management method, characterized in that: The method comprises: Build a blockchain for comprehensive management of product batch information, and add MES servers, DNC servers, CNC machine tools, quality inspection servers, traceability query and AI verification servers to the blockchain; The accounting nodes in the blockchain interact with the MES server, DNC server, CNC machine tool and quality inspection server respectively to record the product information of each product, including: production batch number, product number, production program version number, CNC machine tool number and quality inspection data; each production batch uniquely corresponds to a production batch number; The traceability query and AI verification server obtains the product information of all products under the same production batch from the blockchain, and uses the quality inspection data of all products under the same production batch to train the AI ​​model, and obtains the batch fingerprint data of the corresponding production batch and its matching AI model; When customers need to trace products, they add the product traceability client to the blockchain, and the product traceability client retrieves all product information and corresponding batch fingerprint data and AI models corresponding to the product number of the product to be traced from the traceability query and AI verification server on the blockchain, and obtains the traceability result based on the obtained product information, corresponding batch fingerprint data, AI model and information of the product to be traced; Based on the acquired product information, the corresponding batch fingerprint data, the AI ​​model and the information of the product to be traced, the traceability result is obtained and the following is executed: Obtain product information of the product to be traced according to the nameplate of the product to be traced; The obtained product information is compared with the product information of the product to be traced one by one. If they are not all consistent, the traceability of the product to be traced fails; If all correspondences are consistent, the following is further performed: using the retrieved AI model to process the quality inspection data of the product to be traced to obtain the batch fingerprint characteristics of the product to be traced; if the batch fingerprint characteristics of the product to be traced match the retrieved batch fingerprint characteristics, the traceability of the product to be traced is successful; otherwise, the traceability of the product to be traced fails; In the traceability query and AI verification server, the AI ​​model is trained in the following ways: Perform feature extraction on all quality inspection data of the same production batch to obtain feature vectors of corresponding quality inspection data; Use the feature vectors of all quality inspection data of the same production batch as positive sample data to train the AI ​​model to learn the common features in the positive sample data; After the AI ​​model training is completed, the common features in the positive sample data extracted by the AI ​​model are used as the batch fingerprint data of the corresponding production batch, and the trained AI model is used as the AI ​​model that matches the batch fingerprint data of the corresponding production batch.

2. The product information tamper-proof management method according to claim 1, characterized in that: The feature vector of the quality inspection data is obtained by performing the following operations: Arrange multiple dimensions involved in the quality inspection standards of products of the same production batch in a fixed manner to form a feature extraction template; For each quality inspection data in the same production batch, feature extraction is performed according to the feature extraction template to obtain the feature vector of the corresponding quality inspection data.

3. The product information tamper-proof management method according to claim 1 or 2, characterized in that: The accounting nodes in the blockchain interact with the MES server to monitor the production process of the product and perform: The MES server broadcasts the production schedule to the blockchain; Each production task in the production schedule includes: a production batch number and a preset product number of the product to be processed; The accounting node receives the records published by the MES server and determines whether the production batch number and product number of each production task are recorded in the historical block. If not, the corresponding production task is marked as the first record and written into the blockchain.

4. The product information tamper-proof management method according to claim 3, characterized in that: The accounting nodes in the blockchain interact with the DNC server to supervise the production process of the product and perform: The DNC server receives the production task from the blockchain, successfully sends the production program of the corresponding production task to the CNC machine tool, and broadcasts the corresponding production task and its production program version number to the blockchain; The accounting node receives the record published by the DNC server. If the accounting node finds that the corresponding production task is recorded in the historical block but the production program version number of the corresponding production task is not recorded, the corresponding production task and its production program version number published by the first DNC ​​server broadcast in the blockchain are marked as the second record and written into the blockchain.

5. The product information tamper-proof management method according to claim 4, characterized in that: The accounting nodes in the blockchain interact with CNC machine tools to monitor the production process of products and perform: The data machine tool processes the product based on the production program and assigns a preset product number to the processed product; then, the production batch number in the corresponding production task, the product number of the processed product, and the data machine tool number of the corresponding production task are broadcast to the blockchain; The accounting node receives the records released by the CNC machine tool, and determines whether the first record and the second record in the history block include the production batch number released by the CNC machine tool and the product number of the processed product. If so, the production batch number in the corresponding production task, the product number of the processed product, and the data machine tool number of the corresponding production task are marked as the third record and written into the blockchain.

6. The product information tamper-proof management method according to claim 5, characterized in that: The accounting nodes in the blockchain interact with the quality inspection server to monitor the production process of the product and perform: The quality inspection server conducts quality inspection on the processed products and obtains the quality inspection data; and broadcasts the production batch number, the product number of the processed product and its quality inspection data to the blockchain; The accounting node receives the records released by the quality inspection server, and determines whether the first record, the second record, and the third record in the history block include the production batch number and the product number of the processed product released by the quality inspection server. If so, the production batch number in the corresponding production task, the product number of the processed product and its quality inspection data are marked as the fourth record and written into the blockchain.

7. A product information tamper-proof management system, characterized in that: The system comprises: The blockchain for comprehensive management of product batch information is used to record the product information of each product, including: production batch number, product number, production program version number, CNC machine tool number and quality inspection data; each production batch uniquely corresponds to a production batch number; The traceability query and AI verification server is used to join the blockchain and obtain the product information of all products under the same production batch from the blockchain, and use the quality inspection data of all products under the same production batch to train the AI ​​model to obtain the batch fingerprint data of the corresponding production batch and its matching AI model; The product traceability client is used to join the blockchain when the customer needs to trace the product. The product traceability client retrieves all product information and the corresponding batch fingerprint data and AI model corresponding to the product number of the product to be traced from the traceability query and AI verification server on the blockchain, and obtains the traceability result based on the obtained product information, the corresponding batch fingerprint data, the AI ​​model and the information of the product to be traced; The system also includes an MES server, a DNC server, a CNC machine tool and a quality inspection server; The MES server, DNC server, CNC machine tool and quality inspection server are added to the blockchain, and interact with the MES server, DNC server, CNC machine tool and quality inspection server through the accounting nodes in the blockchain to supervise the production process of the products and record the production batch number, product number, production program version number, CNC machine tool number and quality inspection data of each product; Based on the acquired product information, the corresponding batch fingerprint data, the AI ​​model and the information of the product to be traced, the traceability result is obtained and the following is executed: Obtain product information of the product to be traced according to the nameplate of the product to be traced; The obtained product information is compared with the product information of the product to be traced one by one. If they are not all consistent, the traceability of the product to be traced fails; If all correspondences are consistent, the following is further performed: using the retrieved AI model to process the quality inspection data of the product to be traced to obtain the batch fingerprint characteristics of the product to be traced; if the batch fingerprint characteristics of the product to be traced match the retrieved batch fingerprint characteristics, the traceability of the product to be traced is successful; otherwise, the traceability of the product to be traced fails; In the traceability query and AI verification server, the AI ​​model is trained in the following ways: Perform feature extraction on all quality inspection data of the same production batch to obtain feature vectors of corresponding quality inspection data; Use the feature vectors of all quality inspection data of the same production batch as positive sample data to train the AI ​​model to learn the common features in the positive sample data; After the AI ​​model training is completed, the common features in the positive sample data extracted by the AI ​​model are used as the batch fingerprint data of the corresponding production batch, and the trained AI model is used as the AI ​​model that matches the batch fingerprint data of the corresponding production batch.

Images