Intelligent factory equipment data analysis management system and method based on big data

Abstract

The invention discloses an intelligent factory equipment data analysis management system and method based on big data, and relates to the technical field of intelligent manufacturing, and the method comprises the steps: collecting factory order parameters and operation state data of factory equipment, and calculating the instantaneous power and energy consumption of the factory equipment based on the operation state data; the factory order parameters comprise a processing technology sequence and a delivery time limit; the method comprises the following steps of: performing structured processing on factory order parameters, converting a delivery period into time window limitation, generating a task framework taking minimization of the delivery period and energy consumption as optimization targets, inputting running state data into a spiking neural network model, and outputting a real-time fault probability of factory equipment by simulating a biological neuron synaptic transmission mechanism; and monitoring the operation state data of the factory equipment in real time, and when a sudden fault is identified, updating the scheduling parameter set. Through a time window constraint mechanism, the delivery delay rate is reduced, and through a pulse quantum twin closed-loop architecture, risk conduction dynamic blocking is realized.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of intelligent manufacturing, in particular to an intelligent factory equipment data analysis management system and method based on big data. BACKGROUND

[0002] Intelligent factory equipment data analysis technology has become the core driving force for the digital transformation of manufacturing industry. In the aspect of scheduling optimization, multi-objective genetic algorithm and mixed integer programming have been widely used in the collaborative optimization of delivery deadline and energy consumption. In the aspect of fault prediction, long short-term memory network and convolutional neural network achieve high accuracy in equipment anomaly detection.

[0003] However, the prior art has the following disadvantages: the traditional fault prediction model is independent of the scheduling system and cannot dynamically map the real-time fault probability to the risk propagation coefficient of the topology network. Although the pulse neural network has the advantage of biological neuron synapse transmission, the output fault probability is not coupled with the physical layout of the equipment, resulting in that the scheduling scheme ignores the spatial risk transmission. SUMMARY

[0004] In view of the above existing problems, the present application is proposed.

[0005] Therefore, the present application provides an intelligent factory equipment data analysis management system and method based on big data to solve the problem of missing risk propagation coefficient caused by decoupling of traditional fault prediction model and scheduling system.

[0006] To solve the above technical problems, the present application provides the following technical solutions: In a first aspect, the present application provides an intelligent factory equipment data analysis management method based on big data, which comprises, Collecting factory order parameters and running state data of factory equipment, and calculating the instantaneous power and energy consumption of the factory equipment based on the running state data; the factory order parameters include processing technology sequence and delivery deadline; Structurally processing the factory order parameters, converting the delivery deadline into a time window limit, generating a task framework with the optimization goal of minimizing the delivery deadline and energy consumption, inputting the running state data into a pulse neural network model, and outputting the real-time fault probability of the factory equipment by simulating the biological neuron synapse transmission mechanism; Based on the pre-set physical layout of the factory equipment and the processing technology sequence, a dynamic topology network graph is constructed, and the real-time fault probability is mapped to the risk propagation coefficient between the factory equipment nodes; Based on the task framework, the real-time fault probability and the risk propagation coefficient between the factory equipment nodes, the scheduling parameter set of the factory equipment is generated by quantum calculation; Load the three-dimensional model of the factory equipment and the physical rules in the digital twin mapping space, simulate the operation of the factory equipment based on the scheduling parameter set, eliminate the scheduling parameter set that violates the physical rules, output the executable scheduling parameter set and compile it into machine instructions, and then send it to the factory equipment to execute the production task; Real-time monitoring of the running state data of the factory equipment, when a sudden failure is identified, the scheduling parameter set is updated.

[0007] As a preferred scheme of the intelligent factory equipment data analysis and management method based on big data, the output factory equipment real-time failure probability has the following specific steps, Based on the factory order parameters, the time window constraints of the processing process sequence are constructed, and the delivery deadline is converted into a binary tuple of the earliest start time and the latest completion time of each process; A binary decision variable and an energy consumption prediction function of the factory equipment are established, and a task framework with a core objective function of delivery deadline violation penalty and energy consumption weighted minimization is constructed; The synaptic weight matrix multiplication operation of the pulse neural network is performed on the running state data, and the dynamic failure probability is output through the membrane potential accumulation and pulse firing mechanism.

[0008] As a preferred scheme of the intelligent factory equipment data analysis and management method based on big data, the output factory equipment real-time failure probability has the following specific steps, The pre-set factory equipment physical layout is analyzed, the factory equipment position information is extracted, and the physical connection relationship between the factory equipment is established according to the adjacent distance threshold; The processing process sequence is analyzed, the process dependence direction and dependence level are identified, the directed process dependence relationship between the factory equipment is constructed and the initial weight is given; The physical connection relationship and the process dependence relationship are merged to generate a dynamic topological network graph.

[0009] As a preferred scheme of the intelligent factory equipment data analysis and management method based on big data, the output factory equipment real-time failure probability has the following specific steps, The process dependence relationship weight and the associated factory equipment real-time failure probability are multiplied to calculate the real-time risk propagation coefficient of the process dependence edge; Based on the physical connection relationship between the factory equipment, the real-time failure probability mean of the adjacent factory equipment is fused to calculate the real-time risk propagation coefficient of the physical connection edge; Update the edge attribute of the dynamic topological network graph, write the real-time risk propagation coefficient of the process dependence edge and the real-time risk propagation coefficient of the physical connection edge into the edge attribute field, and in the multiple connection edge scene, take the maximum value of the risk propagation coefficient to cover the old value; A periodic updating mechanism is configured to set a refresh condition to continuously update a real-time risk propagation coefficient of the dynamic topology network graph.

[0010] As a preferred scheme of the intelligent factory equipment data analysis management method based on big data, the generated factory equipment scheduling parameter set has the following specific steps, The binary decision variables and multi-objective optimization functions of the task framework are extracted, and the real-time risk propagation coefficient of the dynamic topology network graph is fused. The binary decision variables are associated with the quantum bit register, the binary decision variables corresponding to the high-risk factory equipment are identified and marked as high-priority bit positions; The multi-objective optimization function is converted into a quantum bit coupling term and a constraint penalty term, and the real-time risk propagation coefficient of the process-dependent edge is used as a weight for weighted fusion to generate a Hamiltonian operator; A parameterized quantum gate layer is constructed, and a phase separator and a mixer are integrated to form a preset hierarchical structure; The superconducting quantum hardware loads the quantum circuit, and multiple measurement samples are obtained to obtain the quantum state probability distribution; The significant probability quantum state is extracted and restored to the factory equipment task allocation table and the process time parameter; The factory equipment task allocation table, the process time parameter, and the optimization confidence index value are output in a structured manner to generate a scheduling parameter set.

[0011] As a preferred scheme of the intelligent factory equipment data analysis management method based on big data, the generated factory equipment scheduling parameter set has the following specific steps, Read the factory equipment three-dimensional model and physical rules, analyze the geometric structure and motion constraints of the factory equipment; Map the task allocation table and process time parameter of the scheduling parameter set to the digital twin mapping space to drive the factory equipment model to perform the task action; Start the physical engine for real-time simulation, and accurately control the action timing and spatial motion of the three-dimensional model based on the time window parameter; Real-time execution of factory equipment collision detection and motion constraint verification to identify events that violate physical rules; Read the conflict event report table and mark the scheduling parameter items that have violation events; Delete the scheduling parameter items with conflict marks to form an executable scheduling parameter set that passes the safety verification; Convert the executable scheduling parameter set into factory equipment operation instructions, timing control instructions, and execution trigger instructions; The execution trigger instruction is transmitted to the factory equipment controller through the industrial bus to trigger the execution of the production task.

[0012] As a preferred embodiment of the big data-based intelligent factory equipment data analysis and management method described in this invention, the specific steps for updating the scheduling parameter set are as follows: By comparing the current operating status data with the real-time dynamic features of the spiking neural network model, abrupt changes and anomalies are identified. The latest data sequence is inferred using the spiking neural network to confirm the generation of the sudden failure event. The status of the faulty factory equipment node is marked and the real-time risk propagation coefficient of the associated edge is reset to the maximum value. The task framework is reconstructed based on the current time, and a new set of scheduling parameters is generated by calling the quantum computing solution engine, and executability is quickly verified. Production tasks are restored by replacing the factory equipment controller execution instruction dataset with an atomic write operation on the industrial bus.

[0013] Secondly, this invention provides a data analysis and management system for intelligent factory equipment based on big data, including an energy consumption calculation module, a fault prediction module, a risk mapping module, a quantum optimization module, an instruction generation module, and a fault detection module. The energy consumption calculation module is used to collect factory order parameters and factory equipment operating status data, and calculate the instantaneous power and energy consumption of the factory equipment based on the operating status data; the factory order parameters include the processing sequence and delivery deadline; The fault prediction module is used to structure the factory order parameters, convert the delivery deadline into a time window limit, generate a task framework with the optimization goal of minimizing the delivery deadline and energy consumption, input the operating status data into the spiking neural network model, and output the real-time fault probability of the factory equipment by simulating the synaptic transmission mechanism of biological neurons. The risk mapping module is used to construct a dynamic topology network diagram based on the physical layout and processing sequence of pre-set factory equipment, and to map the real-time failure probability into the risk propagation coefficient between factory equipment nodes. The quantum optimization module is used to generate a set of scheduling parameters for factory equipment by solving for the task framework, real-time failure probability and risk propagation coefficient between factory equipment nodes through quantum computing. The instruction generation module is used to load the three-dimensional model and physical rules of the factory equipment in the digital twin mapping space, simulate the operation of the factory equipment based on the scheduling parameter set, eliminate the scheduling parameter set that violates the physical rules, output the executable scheduling parameter set and compile it into machine instructions, and send it to the factory equipment to execute production tasks. The fault detection module is used to monitor the operating status data of factory equipment in real time, and update the scheduling parameter set when a sudden fault is detected.

[0014] In a third aspect, the present application provides a computer device comprising a memory and a processor, wherein the memory stores a computer program, and wherein the computer program, when executed by the processor, implements any step of the method for analyzing and managing equipment data of an intelligent factory based on big data according to the first aspect of the present application.

[0015] In a fourth aspect, the present application provides a computer-readable storage medium storing a computer program, wherein the computer program, when executed by a processor, implements any step of the method for analyzing and managing equipment data of an intelligent factory based on big data according to the first aspect of the present application.

[0016] The present application has the following advantages: the time window constraint converts the delivery deadline into a process-level executable boundary, solves the problem of disconnection between the delivery deadline and equipment capacity in traditional scheduling, reduces the delivery delay rate, captures transient risk fluctuations through periodic refreshing, makes the scheduling scheme avoid high-risk plant equipment combinations in real time, realizes dynamic correction of failure probability through biological neuron mechanism, and reduces response delay and false alarm rate compared with the LSTM model. BRIEF DESCRIPTION OF DRAWINGS

[0017] In order to more clearly illustrate the technical solutions of the embodiments of the present application, the following will briefly introduce the drawings needed in the embodiment description. Obviously, the drawings in the following description are only some embodiments of the present application, and other drawings can be obtained by those skilled in the art without creative labor.

[0018] Fig. 1 It is a flowchart of the method for analyzing and managing equipment data of an intelligent factory based on big data.

[0019] Fig. 2 It is a module diagram of the system for analyzing and managing equipment data of an intelligent factory based on big data.

[0020] Fig. 3 It is a flowchart of core data processing and optimization.

[0021] Fig. 4 It is a flowchart of instruction execution and dynamic updating. DETAILED DESCRIPTION

[0022] In order to make the above-mentioned purposes, features and advantages of the present application more obvious and easy to understand, the specific embodiments of the present application will be described in detail below with reference to the drawings.

[0023] In the following description, numerous specific details are set forth in order to provide a thorough understanding of the present application. However, it will be apparent to one skilled in the art that the present application can be practiced without the specific details set forth in this description. In other instances, well-known methods, procedures, components, and circuits have not been described in detail so as not to unnecessarily obscure aspects of the present application.

[0024] It should also be noted that, as used in the specification and in the claims, the article "a", "an", or "the" is intended to mean that there are one or more of the features or elements. As used in this specification and the claims, the term "or" is intended to mean an inclusive "or" rather than an exclusive "or". That is, unless specified otherwise, or clear from the context, the designation "X employs A or B" means that X employs A or B or both A and B. In addition, the articles "a", "an", and "the" are intended to mean that there are one or more (for example, one) of the feature or element in a claim.

[0025] Reference will now be made to the drawings, in which Figs. 1-4 For one embodiment of the present application, the embodiment provides a big data based intelligent factory equipment data analysis management method, comprising the following steps: S1: Collecting factory order parameters and running state data of factory equipment, and calculating instantaneous power and energy consumption of factory equipment based on running state data; the factory order parameters include processing technology sequence and delivery deadline.

[0026] S1.1: Extracting factory order parameters from the manufacturing order database through a structured query operation, including processing technology sequence and delivery deadline.

[0027] Performing a structured query operation in the manufacturing order database, filtering order record within the current production date through date range in the manufacturing order database, reading stored processing technology sequence field and delivery deadline field from the order record, exporting the process name array (maintaining the original arrangement of process steps) as the processing technology sequence according to the flow order of the processing technology sequence field, and parsing the delivery deadline field into the cutoff time node in coordinated universal time format as the delivery deadline.

[0028] S1.2: Collecting running state data of factory equipment in real time based on standard industrial communication protocol, attaching time stamp label and factory coordinate system physical location label to the collected running state data, and obtaining energy consumption through period integration.

[0029] Modbus TCP protocol is selected as a standard industrial communication protocol, a communication device supporting Modbus TCP protocol is installed on each factory equipment, the communication device is connected with the factory equipment controller, the current and voltage of the factory equipment are acquired in real time, the current and voltage are measured discretely through fixed frequency, the instantaneous values of the current and voltage are obtained, the obtained instantaneous values of the current and voltage are defined as sampling points, each sampling point is marked with a time stamp (minimum accuracy 0.001 seconds), and the spatial position coordinates of the factory equipment are mapped with reference to the factory building reference coordinate system, the sampling points with additional time stamp and factory coordinate system label are obtained, then the current and voltage are calculated to obtain instantaneous power, the start and stop time of the factory equipment is determined based on the current and voltage as the integral boundary of instantaneous power, and the energy consumption of the factory equipment is obtained through integral operation of the instantaneous power.

[0030] S1.3: encapsulate the factory order parameters, running state data and energy consumption into a structured transmission unit, and transmit to the processing end through industrial Ethernet.

[0031] A lightweight data container containing a factory order parameter layer, a running state data layer and an energy consumption layer is constructed through a JSON format, the processing technology sequence and the delivery deadline are written in the factory order parameter layer, the current, voltage, time stamp and spatial position coordinates are written in the running state data layer, and the instantaneous power and energy consumption are written in the energy consumption layer, the lightweight data container is generated into a structured transmission unit in accordance with the JavaScript Object Notation format, is transmitted to the pulse neural network processing end through industrial Ethernet, and multi-source information integration is realized.

[0032] S2: structurally process the factory order parameters, convert the delivery deadline into a time window limit, generate a task framework with the optimization objectives of minimizing the delivery deadline and energy consumption, input the running state data into a pulse neural network model, output the real-time failure probability of the factory equipment through simulation of the biological neuron synapse transmission mechanism.

[0033] S2.1: construct a time window constraint of the processing technology sequence based on the factory order parameters, and convert the delivery deadline into a binary group of the earliest start time and the latest completion time of each process.

[0034] The nominal man-hour field stored in the manufacturing work order database is extracted as the process standard processing time, and the total time consumption from the first process to the last process is calculated according to the process standard processing time and the processing process sequence. The delivery deadline in the factory order parameter is extracted and converted into the cutoff time node in the coordinated universal time format, and then from the cutoff time node, the standard processing time of each process is subtracted in reverse order, to obtain the latest completion time of each process. The earliest start time of each process is obtained by subtracting the standard processing time of each process from the latest completion time of each process, and finally the earliest start time and the latest completion time of each process are formed as a binary tuple as the time window constraint of the processing process sequence.

[0035] S2.2: Establish a binary decision variable and a factory equipment energy consumption prediction function, and construct a task framework with delivery deadline violation penalty and energy consumption weighted minimization as the core objective function.

[0036] Based on the binary tuple, a binary decision variable is defined to indicate the allocation relationship between the process and the factory equipment. Based on the factory order parameters, running state data and energy consumption stored in the lightweight data container, a factory equipment energy consumption prediction function is constructed to output the factory equipment energy consumption prediction value. A multi-objective optimization function is constructed by the delivery deadline violation penalty term (linear economic penalty for process overdue delivery) and the factory equipment energy consumption prediction value. The processing process sequence constraint (irreversible and non-adjustable execution order rule that the process must follow in the manufacturing process) and the exclusive condition (at most one process is allocated to the same factory equipment at the same time) are taken as constraint terms, and the binary decision variable and the multi-objective optimization function form a complete task framework.

[0037] Wherein, the expression of the factory equipment energy consumption prediction function is:

[0038] In the formula, is the energy consumption prediction value, is the instantaneous power of the th sampling point, is the total number of sampling points, is the sampling time interval, is the no-load power of the factory equipment, is the no-load time of the factory equipment, is the sampling point serial number index.

[0039] The expression of the multi-objective optimization function is:

[0040] In the formula, is the optimization target output, is the violation penalty term coefficient (unit: yuan per hour) is the total number of processes in the work order, is an index variable of the process, is the completion time of the th process, is the delivery time of the th process, is the time exceeding the delivery deadline, is the cost coefficient of energy consumption (unit: yuan per kilowatt hour), is an index of the factory equipment, is the total number of factory equipment, is the energy consumption prediction value of the th factory equipment.

[0041] S2.3: Perform synaptic weight matrix multiplication operation of the pulse neural network on the running state data, and output the dynamic failure probability through the membrane potential accumulation and pulse firing mechanism.

[0042] Input the running state data into the input layer of the pulse neural network, output the pulse sequence with space-time coding, load the synaptic weight matrix from the local persistent storage medium, perform convolution operation on the synaptic weight matrix and the pulse sequence to obtain the post-synaptic current signal, input the current signal into the membrane potential dynamic equation and accumulate the weighted cumulative membrane potential, when the cumulative membrane potential exceeds the membrane potential threshold (determined according to the pulse frequency statistical distribution of the historical failure data of the factory equipment), the output layer obtains a binary pulse sequence, and the binary pulse sequence is calculated through a fixed time window to obtain the pulse firing frequency, and the pulse firing frequency is mapped to the real-time failure probability of the factory equipment through the Sigmoid activation function.

[0043] S3: Based on the pre-set physical layout of the factory equipment and the processing process sequence, a dynamic topology network graph is constructed, and the real-time failure probability is mapped to the risk propagation coefficient between the nodes of the factory equipment.

[0044] S3.1: Analyze the pre-set physical layout of the factory equipment, extract the position information of the factory equipment, and establish the physical connection relationship between the factory equipment according to the adjacent distance threshold.

[0045] Read the factory equipment physical layout JSON file to obtain the factory equipment position coordinate list, traverse the factory equipment position coordinate list to generate all unordered factory equipment pairs, calculate the Euclidean distance of unordered factory equipment pairs to generate a distance matrix, set the adjacent distance threshold (set according to the maximum value of the physical contour size of the factory equipment, the core function interaction range and the standard safety distance) and filter the unordered factory equipment pairs in the distance matrix less than the adjacent distance threshold, obtain the set of all factory equipment pairs that meet the physical connection condition and assign a unique connection identifier and mark the physical connection type, create a connection relationship table to store the factory equipment through the manufacturing order database, including the connection identifier column and the physical connection type column, traverse the set of factory equipment pairs, take each factory equipment pair and the physical connection type mark as a record, and insert it into the connection identifier column and the physical connection type column. The final complete record table is used as the physical connection relationship.

[0046] S3.2: Analyze the processing process sequence, identify the process dependency direction and dependency level, and construct the directed process dependency relationship between the factory equipment and assign an initial weight.

[0047] Read the processing process sequence to extract the process node, the dependency direction between processes, the factory equipment task allocation table and the process dependency level, map the process node to the factory equipment based on the factory equipment task allocation table to obtain the process dependency relationship between the factory equipment, and through the mapping operation of the factory equipment task allocation table, convert the process node pair in the process dependency relationship to the factory equipment pair. At the same time, the dependency direction between processes and the process dependency level are directly bound to the factory equipment pair to generate a directed process dependency relationship between the factory equipment containing the source factory equipment, the target factory equipment and the initial weight value.

[0048] S3.3: Merge the physical connection relationship and the process dependency relationship to generate a dynamic topology network graph containing factory equipment nodes, physical connection edges and process dependency edges.

[0049] Read the physical connection relationship to extract the factory equipment pair and the physical connection type, receive the source factory equipment, the target factory equipment and the initial weight value of the directed process dependency relationship between the factory equipment, create a blank graph data structure instance, traverse the factory equipment registered as an independent node in the graph data structure, traverse the physical connection relationship to add undirected physical connection edges one by one and mark the connection type as the physical connection type in the edge attribute, traverse the directed process dependency relationship between the factory equipment, add the process dependency edge from the source factory equipment to the target factory equipment in the blank graph data structure instance and set the initial weight value and the process connection type mark in the edge attribute. Finally, the initial structure of the dynamic topology network graph with complete factory equipment nodes and edge attributes is formed.

[0050] S3.4: Write the real-time failure probability into the corresponding factory equipment node attribute field to complete the dynamic topology network graph state update.

[0051] receiving the real-time fault probability of the pulse neural network model output, parsing the factory equipment identifier and fault probability key-value pair in the real-time fault probability, locating the corresponding factory equipment node in the dynamic topology network graph according to the factory equipment identifier, writing the real-time fault probability into the fault probability storage address in the factory equipment node attribute field, and generating the dynamic topology network graph containing the latest factory equipment node state.

[0052] S3.5: Perform a product operation on the process dependency relationship weight and the associated real-time fault probability of the factory equipment, and calculate the real-time risk propagation coefficient of the process dependency edge.

[0053] S3.5: Perform a product operation on the process dependency relationship weight and the associated real-time fault probability of the factory equipment, and calculate the real-time risk propagation coefficient of the process dependency edge.

[0054] S3.6: Based on the physical connection relationship between factory equipments, fuse the real-time fault probability mean of adjacent factory equipments, and calculate the real-time risk propagation coefficient of the physical connection edge.

[0055] S3.6: Based on the physical connection relationship between factory equipments, fuse the real-time fault probability mean of adjacent factory equipments, and calculate the real-time risk propagation coefficient of the physical connection edge.

[0056] S3.7: Update the edge attribute of the dynamic topology network graph, write the real-time risk propagation coefficient of the process dependency edge and the real-time risk propagation coefficient of the physical connection edge into the edge attribute field, and in the case of multiple connection edges, take the maximum value of the risk propagation coefficient to cover the old value.

[0057] Traverse all process-dependent edges and physical connection edges in the dynamic topology network graph. Write the real-time risk propagation coefficient of each process-dependent edge and physical connection edge into the risk propagation coefficient storage address in the corresponding edge attribute field to overwrite the original value. Retrieve multiple connection edges existing in the same factory equipment pair (such as physical connection edges and process-dependent edges coexisting). Compare the risk propagation coefficients of all process-dependent edges and physical connection edges among the multiple connection edges and record the maximum value. Rewrite the maximum value into the corresponding attribute field to achieve a unified overwrite of the final value. Generate a dynamic topology network graph with updated edge attributes carrying the real-time risk propagation coefficient.

[0058] S3.8: Configure a periodic update mechanism and set refresh conditions to continuously update the real-time risk propagation coefficient of the dynamic topology network diagram.

[0059] In the dynamic topology network graph after edge attribute updates, initialize a fixed interval timer and set a default interval parameter (e.g., 250 milliseconds). Continuously execute the detection that the fixed interval timer reaches the interval parameter and the real-time fault probability change rate exceeds the 5% change rate threshold (set according to the absolute value of the fault probability change rate in industrial diagnostic standards). When either condition is met, trigger the risk propagation coefficient recalculation process start signal, execute the risk coefficient calculation and attribute update operation for process-dependent edges and physically connected edges, reset the timer after completion, and start the next waiting cycle. Continuously refresh the real-time risk propagation coefficient of the dynamic topology network graph.

[0060] S4: Based on the task framework, real-time failure probability, and risk propagation coefficient between factory equipment nodes, the scheduling parameter set of factory equipment is generated by solving through quantum computing.

[0061] S4.1: Extract the binary decision variables and multi-objective optimization functions of the task framework, integrate the real-time risk propagation coefficients of the dynamic topology network graph, and generate the quantum computing input data volume.

[0062] The mathematical structure of the binary decision variables and multi-objective optimization function stored in the task framework is read, the real-time risk propagation coefficient key-value pairs output by the dynamic topology network graph are loaded, the binary decision variables are mapped to a sequence of qubit identifiers, the product of the predicted energy consumption of factory equipment and the real-time risk propagation coefficient in the multi-objective optimization function is defined as a weighted term, the penalty term for delivery deadline violation is converted into a constraint term (the delivery deadline must be met), and the quantum computing input data volume is generated by uniformly converting the multi-objective optimization function, constraint term and weighted term into the quadratic polynomial form of qubits (such as the Ising model) and performing normalization processing.

[0063] S4.2: Associate binary decision variables with a qubit register to identify binary decision variables corresponding to high-risk factory equipment and mark high-priority bits.

[0064] According to the binary decision variable, a quantum bit mapping table is generated through a quantum conversion process, the corresponding relationship between the binary decision variable and the quantum bit in the quantum bit mapping table is loaded, the real-time failure probability of the factory equipment node in the dynamic topology network graph is traversed, the real-time failure probability is sorted in descending order, the factory equipment with the real-time failure probability in the top percentage (such as the top 15%) is selected as the high-risk factory equipment, the binary decision variable in the quantum bit mapping table whose factory equipment index number belongs to the high-risk factory equipment is identified, the priority attribute value of the quantum bit corresponding to the binary decision variable in the quantum register mapping record is set to high priority, and a quantum bit mapping relationship table with a high priority mark is generated.

[0065] S4.3: Convert the multi-objective optimization function into a quantum bit coupling term and a constraint penalty term, and weight and fuse them with the real-time risk propagation coefficient of the process-dependent edge as the weight to generate a Hamiltonian operator.

[0066] Load the mathematical structural expression of the multi-objective optimization function stored in the task framework, parse the algorithm of the energy consumption term and identify the binary decision variable product coupling structure therein, convert the binary decision variable product term in the energy consumption term into a quantum bit coupling term, parse the constraint logic of the deadline violation penalty term to construct a slack variable penalty term (allowing violation but punishing) as a quantum bit constraint penalty term, traverse the real-time risk propagation coefficient key-value pair of the dynamic topology network graph to retrieve the real-time risk propagation coefficient matching the process-dependent edge, multiply the real-time risk propagation coefficient of the process-dependent edge into the quantum bit coupling term as a weight factor to generate a weighted coupling expression, and integrate all quantum bit coupling terms and constraint penalty terms to generate a Hamiltonian operator.

[0067] S4.4: Construct a parameterized quantum gate layer, integrate a phase separator and a mixer to form a preset hierarchical structure.

[0068] Load the quantum bit mapping table to obtain the quantum bit register address information and the high priority marked bit, receive the quantum bit coupling term and the constraint penalty term in the Hamiltonian operator, initialize the quantum circuit according to the preset hierarchical structure parameters (for example, the preset hierarchical number is three layers), construct a Hadamard gate in the initial layer of the quantum circuit to act on all quantum bits to generate a uniform superposition state, apply a quantum phase separation calculation unit, control the Z-axis rotation operation of the quantum bit by adjusting the angle parameter of the unit (for example, using 0.8π radians) to realize task cost calculation, simultaneously apply a quantum mixing calculation unit, drive the X-axis rotation operation of the quantum bit by adjusting the angle parameter of the unit (for example, using 0.6π radians) to perform space exploration, and repeat the construction process to form the complete structure of the parameterized quantum gate layer when the preset number of times (for example, 3 times) is reached.

[0069] S4.5: Load the quantum circuit in the superconducting quantum hardware, run multiple measurement sampling to obtain the quantum state probability distribution.

[0070] Receive the parameterized quantum gate layer and convert it into a set of physical microwave pulse signals through the superconducting quantum hardware control interface, start the initial calibration process of the superconducting quantum processor to verify the phase coherence of the quantum bits, guide the set of physical microwave pulse signals into the quantum bit control line to execute the quantum gate operation sequence, apply the measurement gate operation to each quantum bit after the quantum circuit is executed to record the binary state value generated by the quantum state collapse, repeat the quantum circuit operation sequence multiple times, accumulate the measurement results, and calculate the normalized quantum state probability distribution based on the number of quantum bits and the measurement error of the quantum hardware.

[0071] S4.6: Extract the significant probability quantum state and restore it to the factory equipment task allocation table and process time parameter.

[0072] Traverse the normalized quantum state probability distribution, filter out the quantum states with probability exceeding the confidence threshold (set dynamically based on the number of quantum bits and the measurement error of the quantum hardware), convert the quantum state to the corresponding binary string (each quantum state corresponds to a string generated by the quantum bit state value), parse each quantum bit state value in the binary string based on the quantum bit mapping table, map the quantum bit state value back to the assignment of the corresponding binary decision variable, extract the factory equipment identifier and task identifier associated with the quantum bit based on the factory equipment index identifier, and calculate the start time and end time of each task based on the process dependency relationship and delivery deadline stored in the task framework. Finally, output the factory equipment task allocation table and process time parameter.

[0073] S4.7: Structured output of factory equipment task allocation table, process time parameter and optimization confidence index value, generate scheduling parameter set.

[0074] Receive the factory equipment task allocation table and read each task allocation record, load the process time parameter and extract the planned start time and planned end time corresponding to each task identifier, obtain the optimization confidence index value obtained by quantum state measurement calculation, associate the factory equipment allocation information, process time parameter and optimization confidence index value according to the task identifier field, and integrate the task identifier, factory equipment identifier, process time parameter and optimization confidence index value into the scheduling parameter set.

[0075] S5: Load the factory equipment three-dimensional model and physical rules in the digital twin mapping space, simulate the operation of the factory equipment based on the scheduling parameter set, eliminate the scheduling parameter set that violates the physical rules, output the executable scheduling parameter set and compile it into machine instructions, and issue it to the factory equipment to execute the production task.

[0076] S5.1: Read the factory equipment three-dimensional model and physical rules, analyze the geometry and motion constraints of the factory equipment.

[0077] Read the GLTF format storage data stream of the factory equipment three-dimensional model in the digital twin mapping space, reconstruct the geometry of the factory equipment by analyzing the grid vertex coordinates and face index data in the factory equipment three-dimensional model, and load the JSON format constraint file of the physical rules, extract the collision body bounding box size parameters, analyze the joint angle upper limit (the relative rotation angle upper limit between the motion components of the factory equipment, defined according to the factory equipment manufacturer's design parameters) and velocity threshold (set according to the factory equipment performance manual calibration parameters) in the motion constraint field of the physical rules, and bind the factory equipment geometry and motion constraint rules to the three-dimensional data structure associated with the unified factory equipment identifier.

[0078] S5.2: Map the task allocation table of the scheduling parameter set and the process time parameters to the digital twin mapping space, and drive the factory equipment model to perform task actions.

[0079] Load the task allocation table stored in the scheduling parameter set, read the factory equipment identifier information associated with each task identifier, extract the process time parameters recorded in the scheduling parameter set to obtain the task start and end time, then search for the factory equipment three-dimensional model matching the factory equipment identifier in the digital twin mapping space, bind the process time parameters to the animation controller driving time axis of the factory equipment three-dimensional model, and trigger the factory equipment three-dimensional model to perform simulated actual production task flow according to the start time.

[0080] S5.3: Start the physical engine real-time simulation, accurately control the action timing and spatial motion of the three-dimensional model based on the time window parameters.

[0081] Set the physical engine simulation timer acceleration factor (such as acceleration factor 10) in the digital twin mapping space, load the animation controller, calculate the action duration of the factory equipment three-dimensional model based on the process time parameters, synchronize the physical engine dynamics constraints of the three-dimensional model, start the physical engine to update the spatial position and posture of the factory equipment three-dimensional model at an update frequency (such as 0.02 seconds), and record the motion trajectory of the factory equipment three-dimensional model in the space-time coordinate system in real time according to the action duration.

[0082] S5.4: Real-time execution of factory equipment collision detection and motion constraint verification, identification of events violating physical rules.

[0083] The collision body bounding box size parameters and velocity threshold of the physical rules are loaded, the real-time motion trajectory of the three-dimensional model is obtained, the minimum Euclidean distance between the adjacent factory equipment colliders in the motion trajectory is calculated, whether the minimum Euclidean distance is less than the safety spacing threshold (set according to the dynamic characteristics parameters of the factory equipment) is compared to identify the collision risk event, the joint angle range in the physical rules is analyzed synchronously (determined according to the upper limit of the joint angle and the factory measurement), whether the joint angle in the motion trajectory exceeds the joint angle range is detected to identify the constraint violation event, and finally the identified collision event and constraint violation event are written into the conflict event report table to complete the record of the violation of the physical rule event.

[0084] S5.5: Read the conflict event report table and mark the scheduling parameter item with the existing violation event.

[0085] The conflict event report table includes the fields of conflict type, time, factory equipment identifier, and deviation amount of safety spacing threshold. The real-time output collision event and constraint violation event record are traversed, the event occurrence time point, associated factory equipment identifier, and deviation amount are extracted for each constraint violation event according to the type, and the complete record row is filled into the corresponding field of the conflict event report table to generate a complete record row. The task identifier of the scheduling parameter set is loaded synchronously, and it is checked whether there is a conflict record with the same factory equipment identifier as the current task in the conflict event report table. If there is, the violation event marker field is written in the task item of the scheduling parameter set. Finally, the scheduling parameter set with violation markers is output.

[0086] S5.6: Delete the scheduling parameter item with the conflict marker to form a safe and executable scheduling parameter set.

[0087] The scheduling parameter set with violation event markers is loaded, a blank executable scheduling parameter set storage structure is created, each task record of the original scheduling parameter set is traversed, and it is detected whether the violation marker field of the current task record is a Boolean true value. If it is false, the complete content of the task record is copied to the executable scheduling parameter set. If it is true, the task record is skipped and not copied. Finally, the executable scheduling parameter set without any conflict marker item is output to complete the safety verification screening.

[0088] S5.7: Convert the executable scheduling parameter set into an instruction data set containing factory equipment operation instructions, timing control instructions, and execution trigger instructions.

[0089] The task record loaded with the executable scheduling parameter set extracts the factory equipment identifier and task type field of each task, queries the factory equipment operation instruction template library according to the task type field to match the corresponding instruction template to generate the factory equipment operation instruction, obtains the start time and end time from the task record and converts them into the start and end timestamp format that can be parsed by the factory equipment controller to generate the timing control instruction, sets the execution trigger instruction to a Boolean true value (because it has passed the conflict screening), and organizes the factory equipment operation instruction, timing control instruction and execution trigger instruction to form the instruction data set.

[0090] S5.8: Transmit the execution trigger instruction to the factory equipment controller through the industrial bus to trigger the execution of the production task.

[0091] Traverse the instruction data set to extract the execution trigger instruction, call the industrial bus driver interface (such as COM1 RS485) to establish a physical connection with the factory equipment controller, send the execution trigger instruction (function code 05 forced to be set) to activate the factory equipment controller to execute the production task, and close the industrial bus connection after confirming that the factory equipment controller returns a success code.

[0092] S6: Real-time monitoring of the running state data of the factory equipment, when a sudden failure is identified, update the scheduling parameter set.

[0093] S6.1: Compare the current running state data with the real-time dynamic characteristic value of the pulse neural network model to identify the mutation anomaly, call the pulse neural network to infer the latest data sequence to confirm the generation of a sudden failure event, mark the fault factory equipment node state and reset the associated edge risk propagation coefficient to the maximum value.

[0094] Extract the latest timestamp running state data, convert it to the real-time dynamic characteristic value of the factory equipment through the pulse neural network model, calculate the absolute change rate of the real-time dynamic characteristic value relative to the factory equipment safety threshold (set according to the factory equipment failure economic loss model and the pulse neural network confidence feature), repeat the calculation of the temperature and current in the latest timestamp running state data, and integrate the absolute change rate relative to the factory equipment safety threshold, the absolute change rate of the temperature and current, when any absolute change rate exceeds the set mutation threshold (determined according to the factory equipment hard failure physical characteristics and the pulse neural network confidence sensitivity) and both absolute change rates reach the warning threshold (determined according to the safety margin of the factory equipment technical specifications and the historical failure data characteristics) at the same time, mark the sudden abnormal event to complete the mutation anomaly identification.

[0095] Extract the continuous running state data stored in the preset time window (e.g., set to 10 seconds) in the running state data, standardize and encode the running state data according to the input format requirements of the spiking neural network model, load the synaptic weight matrix of the spiking neural network model, execute the calculation process of the spiking neuron synaptic transmission mechanism to output the real-time fault probability, continuously monitor whether the real-time fault probability exceeds the safety threshold and the time duration, and when the real-time fault probability continuously exceeds the safety threshold for a time duration of the shortest confirmation time (500 ms), generate a burst fault confirmation event and output it to the fault response process, otherwise reset the time duration and continue monitoring.

[0096] Load the fault factory equipment identifier contained in the burst fault confirmation event, retrieve the factory equipment node matching the factory equipment identifier in the dynamic topology network graph, update the state attribute field of the factory equipment node to the "fault shutdown" state, and simultaneously traverse all process dependency edges and physical connection edges connected to the fault factory equipment node in the dynamic topology network graph, forcibly update the real-time risk propagation coefficient of each process dependency edge and physical connection edge to the maximum value, and generate an updated dynamic topology network graph.

[0097] S6.2: Reconstruct the task framework based on the current time and call the quantum computing solving engine to generate a new set of scheduling parameters and quickly verify the executability.

[0098] Get the current high-precision timestamp as the reconstruction reference time, load the delivery deadline of the unfinished tasks in the task framework, reset the planned start time lower limit of all unstarted tasks to the current reference time, recalculate the updated time window end time based on the original delivery deadline to form the modified time window parameters, replace the original time window parameters with the modified time window parameters in the task framework to form the reconstructed task framework, call the quantum computing solving engine to input the reconstructed task framework, the current real-time fault probability, and the updated dynamic topology network graph, and execute quantum computing to generate a new set of scheduling parameters.

[0099] Load the factory equipment three-dimensional model and physical rules in the digital twin mapping space, extract the task change set (only containing affected factory equipment task items) associated with the fault factory equipment in the new scheduling parameter set, retrieve the factory equipment three-dimensional model based on the task identifier of the task change set, drive the three-dimensional model to perform actions according to the task time window parameters of the new scheduling parameter set, start accelerated simulation (10 times real-time speed) to perform collision detection and motion constraint verification only for the task change set related factory equipment, generate physical rule verification results exclusive to the task change set, and integrate the verified task items to form a locally executable scheduling parameter set.

[0100] S6.3: Replace the factory equipment controller execution instruction data set through an industrial bus atomic write operation to restore production tasks.

[0101] Load local executable scheduling parameter set extraction task identifier, call machine instruction compilation flow to convert local executable scheduling parameter set into triple sequence, establish communication connection with target factory equipment controller through industrial bus driver interface, prepare industrial bus atomic write operation transaction (Modbus function code 16) execution instruction data set, send atomic write command to factory equipment controller instruction storage area start address, verify factory equipment controller return operation success response code, immediately send execution trigger instruction to restore production task.

[0102] The embodiment also provides an intelligent factory equipment data analysis management system based on big data, comprising: The energy consumption calculation module, the fault prediction module, the risk mapping module, the quantum optimization module, the instruction generation module and the fault detection module, the energy consumption calculation module is used for collecting factory order parameters and running state data of the factory equipment, and the instantaneous power and the energy consumption of the factory equipment are calculated based on the running state data; the factory order parameters include processing process sequence and delivery deadline; the fault prediction module is used for structuring the factory order parameters, converting the delivery deadline into a time window limit, generating a task framework with the optimization target of minimizing the delivery deadline and the energy consumption, inputting the running state data into a pulse neural network model, outputting the real-time fault probability of the factory equipment by simulating the synaptic transmission mechanism of biological neurons; the risk mapping module is used for constructing a dynamic topological network graph based on the pre-set physical layout of the factory equipment and the processing process sequence, and mapping the real-time fault probability into the risk propagation coefficient between the factory equipment nodes; the quantum optimization module is used for generating the scheduling parameter set of the factory equipment by quantum calculation based on the task framework, the real-time fault probability and the risk propagation coefficient between the factory equipment nodes; the instruction generation module is used for loading the three-dimensional model and the physical rules of the factory equipment in the digital twin mapping space, simulating the running of the factory equipment based on the scheduling parameter set, eliminating the scheduling parameter set that violates the physical rules, outputting the executable scheduling parameter set and compiling it into machine instructions, and delivering the machine instructions to the factory equipment to execute the production task; and the fault detection module is used for monitoring the running state data of the factory equipment in real time, and updating the scheduling parameter set when a sudden fault is identified.

[0103] The embodiment also provides a computer device suitable for the case of the intelligent factory equipment data analysis management method based on big data, comprising: a memory and a processor; the memory is used for storing computer executable instructions, and the processor is used for executing the computer executable instructions to realize the intelligent factory equipment data analysis management method based on big data as described in the above embodiment.

[0104] The computer device can be a terminal, and the computer device includes a processor, a memory, a communication interface, a display screen and an input device connected by a system bus. The processor of the computer device is configured to provide computing and control capabilities. The memory of the computer device includes a non-volatile storage medium and an internal memory. The non-volatile storage medium stores an operating system and a computer program. The internal memory provides an environment for running the operating system and the computer program in the non-volatile storage medium. The communication interface of the computer device is configured to perform wired or wireless communication with an external terminal. The wireless communication can be achieved by WIFI, an operator network, NFC (Near Field Communication) or other technologies. The display screen of the computer device can be a liquid crystal display screen or an electronic ink display screen. The input device of the computer device can be a touch layer overlaid on the display screen, or a key, trackball or touchpad arranged on the shell of the computer device, or an external keyboard, touchpad or mouse, etc.

[0105] The embodiment also provides a storage medium having a computer program stored thereon, the program being executed by a processor to implement the method for analyzing and managing equipment data of an intelligent factory based on big data. The storage medium can be implemented by any type of volatile or non-volatile storage device or a combination thereof, such as a static random access memory (SRAM), an electrically erasable programmable read-only memory (EEPROM), an erasable programmable read-only memory (EPROM), a programmable read-only memory (PROM), a read-only memory (ROM), a magnetic memory, a flash memory, a magnetic disk or an optical disk.

[0106] In summary, the present application converts the delivery deadline into a process-level executable boundary by time window constraints, solves the problem that the delivery deadline is not consistent with the equipment capacity in the traditional scheduling, reduces the delivery delay rate, captures the instantaneous risk fluctuations through periodic refreshing, makes the scheduling scheme avoid the high-risk plant equipment combination in real time, realizes dynamic correction of failure probability through the biological neuron mechanism, and reduces the response delay and false alarm rate of the LSTM model.

[0107] It should be noted that the above examples are only used to illustrate the technical solutions of the present application but not limit the present application. Although the present application is described in detail with reference to the preferred embodiments, those skilled in the art should understand that the technical solutions of the present application can be modified or equivalently replaced, without departing from the spirit and scope of the technical solutions of the present application, which should be covered in the scope of the claims of the present application.

Claims

1. A big data-based intelligent factory equipment data analysis management method, characterized in that: Comprising, Collecting factory order parameters and running state data of factory equipment, and calculating instantaneous power and energy consumption of the factory equipment based on the running state data; the factory order parameters include processing technology sequence and delivery deadline; Structurally processing the factory order parameters, converting the delivery deadline into a time window limit, generating a task framework with the optimization goal of minimizing the delivery deadline and energy consumption, inputting the running state data into a spiking neural network model, and outputting real-time failure probability of the factory equipment by simulating biological neuron synapse transmission mechanism; Based on the pre-set physical layout of the factory equipment and the processing technology sequence, a dynamic topology network graph is constructed, and the real-time failure probability is mapped to the risk propagation coefficient between the nodes of the factory equipment; Based on the task framework, the real-time failure probability, and the risk propagation coefficient between the nodes of the factory equipment, a scheduling parameter set of the factory equipment is generated by quantum calculation; Loading the three-dimensional model and physical rules of the factory equipment in the digital twin mapping space, simulating the operation of the factory equipment based on the scheduling parameter set, eliminating the scheduling parameter set that violates the physical rules, outputting the executable scheduling parameter set and compiling it into machine instructions, and issuing it to the factory equipment to execute the production task; Real-time monitoring of the running state data of the factory equipment, updating the scheduling parameter set when a sudden failure is identified. 2.The big data based intelligent factory equipment data analysis management method of claim 1, wherein: The output real-time failure probability of the factory equipment, the specific steps are as follows, Based on the factory order parameters, the time window constraints of the processing technology sequence are constructed, and the delivery deadline is converted into a binary tuple of the earliest start time and the latest completion time of each process; Establishing a binary decision variable and a factory equipment energy consumption prediction function, and constructing a task framework with the core objective function of delivery deadline violation penalty and energy consumption weighted minimization; Performing synaptic weight matrix multiplication operation of spiking neural network on the running state data, and outputting dynamic failure probability through membrane potential accumulation and pulse emission mechanism.

3. The big data based smart factory equipment data analysis management method of claim 1, wherein: The specific steps of constructing the dynamic topology network graph are as follows, Analyzing the pre-set physical layout of the factory equipment, extracting the position information of the factory equipment, and establishing the physical connection relationship between the factory equipment according to the adjacent distance threshold; Analyzing the processing technology sequence, identifying the process dependency direction and level, constructing the directed process dependency relationship between the factory equipment and assigning the initial weight; Merging the physical connection relationship and the process dependency relationship to generate a dynamic topology network graph.

4. The big data based smart factory equipment data analysis management method of claim 1, wherein: The specific steps of mapping the real-time failure probability to the risk propagation coefficient between the nodes of the factory equipment are as follows, Performing product operation on the process dependency relationship weight and the associated real-time failure probability of the factory equipment to calculate the real-time risk propagation coefficient of the process dependency edge; Based on the physical connection relationship between the factory equipment, the real-time failure probability mean of the adjacent factory equipment is fused to calculate the real-time risk propagation coefficient of the physical connection edge; Updating the edge attribute of the dynamic topology network graph, writing the real-time risk propagation coefficient of the process dependency edge and the real-time risk propagation coefficient of the physical connection edge into the edge attribute field, and taking the maximum value of the risk propagation coefficient to cover the old value in the multiple connection edge scenario; Configuring a periodic updating mechanism and setting a refresh condition to continuously update the real-time risk propagation coefficient of the dynamic topology network graph.

5. The big data based smart factory equipment data analysis management method of claim 1, wherein: The specific steps of generating the scheduling parameter set of the factory equipment are as follows, The binary decision variable and the multi-objective optimization function of the extraction task framework are fused with the real-time risk propagation coefficient of the dynamic topology network graph; The binary decision variable is associated with the quantum bit register, and the binary decision variable corresponding to the high-risk plant equipment is identified and marked as a high-priority bit; The multi-objective optimization function is converted into a quantum bit coupling term and a constraint penalty term, and the real-time risk propagation coefficient of the process-dependent edge is used as a weight for weighted fusion to generate a Hamiltonian operator; A parameterized quantum gate layer is constructed, and a phase separator and a mixer are integrated to form a preset hierarchical structure; The superconducting quantum hardware loads the quantum circuit, and multiple measurement samples are obtained to obtain the quantum state probability distribution; The significant probability quantum state is extracted and restored to the plant equipment task allocation table and the process time parameter; The plant equipment task allocation table, the process time parameter, and the optimization confidence index value are structured and output to generate a scheduling parameter set.

6. The big data based smart factory equipment data analysis management method of claim 1, wherein: The execution of the production task is as follows, Read the plant equipment three-dimensional model and physical rules, analyze the plant equipment geometric structure and motion constraints; Map the task allocation table and process time parameter of the scheduling parameter set to the digital twin mapping space to drive the plant equipment model to execute the task action; Start the physical engine for real-time simulation, and accurately control the three-dimensional model action timing and spatial motion based on the time window parameter; Real-time plant equipment collision detection and motion constraint verification to identify events that violate physical rules; Read the conflict event report table and mark the scheduling parameter items that have violation events; Delete the scheduling parameter items with conflict marks to form an executable scheduling parameter set that passes the safety verification; Convert the executable scheduling parameter set into plant equipment operation instructions, timing control instructions, and execution trigger instructions; Transmit the execution trigger instructions to the plant equipment controller through the industrial bus to trigger the execution of the production task.

7. The big data based smart factory equipment data analysis management method of claim 1, wherein: The update of the scheduling parameter set is as follows, Compare the current running state data with the real-time dynamic characteristic value of the pulse neural network model to identify sudden abnormalities, call the pulse neural network to infer the latest data sequence to confirm the sudden failure event generation, mark the fault plant equipment node state and reset the real-time risk propagation coefficient of the associated edge to the maximum value; Reconstruct the task framework based on the current time and call the quantum computing solution engine to generate a new scheduling parameter set and quickly verify its executability; Replace the plant equipment controller execution instruction data set through the industrial bus atomic write operation to restore the production task.

8. A big data-based intelligent factory equipment data analysis management system based on any one of the big data-based intelligent factory equipment data analysis management methods of claims 1-7. It includes an energy consumption calculation module, a fault prediction module, a risk mapping module, a quantum optimization module, an instruction generation module, and a fault detection module. The energy consumption calculation module is used to collect plant order parameters and plant equipment running state data, and calculate the instantaneous power and energy consumption of the plant equipment based on the running state data. The plant order parameters include processing process sequence and delivery deadline. The fault prediction module is used to structure the plant order parameters, convert the delivery deadline into a time window limit, generate a task framework with the optimization objectives of minimizing the delivery deadline and energy consumption, input the running state data into a pulse neural network model, output the real-time failure probability of the plant equipment by simulating the biological neuron synapse transmission mechanism, and output the real-time failure probability of the plant equipment. The risk mapping module is configured to construct a dynamic topology network diagram based on the physical layout and the processing process sequence of the preset factory equipment, and map the real-time failure probability to a risk propagation coefficient between nodes of the factory equipment. The quantum optimization module is configured to solve and generate a set of scheduling parameters of the factory equipment by quantum calculation based on a task framework, the real-time failure probability, and the risk propagation coefficient between the nodes of the factory equipment. The instruction generation module is configured to load a three-dimensional model and physical rules of the factory equipment in a digital twin mapping space, simulate running of the factory equipment based on the set of scheduling parameters, eliminate the set of scheduling parameters that violate the physical rules, output an executable set of scheduling parameters, compile the executable set of scheduling parameters into machine instructions, and issue the machine instructions to the factory equipment to execute a production task. The failure detection module is configured to monitor running state data of the factory equipment in real time, and update the set of scheduling parameters when a sudden failure is identified. 9.A computer device, comprising a memory and a processor, wherein the memory stores a computer program, and the computer device is characterized in that: The processor executes the computer program to implement the steps of the method for analyzing and managing data of intelligent factory equipment based on big data according to any one of claims 1-7.

10. A computer readable storage medium having stored thereon a computer program, characterized in that: The computer program is executed by the processor to implement the steps of the method for analyzing and managing data of intelligent factory equipment based on big data according to any one of claims 1-7.

Images