A method of processing knowledge, a computing device, a storage medium

Abstract

The application discloses a knowledge processing method, comprising the following steps: verifying the to-be-processed information by using a verification model in response to the to-be-processed information sent by a user, obtaining knowledge object data, verification data and applicable boundary data, and obtaining new paradigm knowledge based on the knowledge object data, the verification data and the applicable boundary data. The knowledge processing method, the computing device and the storage medium provided by the application can generate extended knowledge or new paradigm knowledge based on to-be-processed information, and improve the standardization and adaptability of knowledge processing.

Description

Technical Field

[0001] This application relates to the field of artificial intelligence, and more specifically, to a knowledge processing method, computing device, and storage medium. Background Technology

[0002] In the era of artificial intelligence, traditional education systems and knowledge acquisition methods face fundamental challenges. Experience, common sense, and theory are being restructured, and old methods are struggling to solve new problems. Current educational paradigms overemphasize the transmission of deterministic knowledge and formalized training, suppressing the instinct for questioning and the spirit of inquiry, leading to cognitive rigidity and an inability to cope with the complex challenges of a rapidly changing environment. Existing knowledge processing methods lack standardized, structured verification and generation mechanisms, making it difficult to optimize or restructure knowledge; the processing methods are insufficiently standardized and lack adaptability. Summary of the Invention

[0003] To address the aforementioned problems, this invention proposes a knowledge processing method, computing device, and storage medium that can generate extended knowledge or new paradigm knowledge based on the information to be processed, thereby improving the standardization and adaptability of knowledge processing.

[0004] This invention provides a knowledge processing method, the method comprising: responding to information to be processed sent by a user, verifying the information to be processed using a verification model, and obtaining knowledge object data, verification data, and applicable boundary data; and obtaining new paradigm knowledge based on the knowledge object data, the verification data, and the applicable boundary data.

[0005] In one embodiment, the step of responding to pending information sent by a user, verifying the pending information using a verification model, and obtaining knowledge object data, verification data, and applicable boundary data includes: performing semantic parsing processing on the pending information using the verification model to obtain knowledge object data; performing system verification processing on the knowledge object data using the verification model to obtain verification data; and performing extended verification processing on the knowledge object data using the verification model to obtain applicable boundary data.

[0006] In one embodiment, the knowledge object data includes conceptual information, logical relationship information, target indicator information, constraint information, and application scenario information; the verification data includes consistency score, validity score, and adaptability score; the verification model performs systematic verification processing on the knowledge object data to obtain verification data, including: the verification model performs logical rule matching and conflict detection processing based on the conceptual information and the logical relationship information to obtain the consistency score; the verification model performs target matching calculation processing based on the target indicator information to obtain the validity score; and the verification model performs environmental simulation analysis processing based on the constraint information and the application scenario information to obtain the adaptability score.

[0007] In one embodiment, the applicable boundary data includes extensibility parameters and demand adaptation parameters. The verification model performs extensibility verification processing on the knowledge object data to obtain the applicable boundary data, including: the verification model constructs module dependencies based on the logical relationship information and performs structural analysis processing to obtain the extensibility parameters; the verification model constructs a target demand model based on the target indicator information and the application scenario information, and performs matching calculation processing between the application scenario information and the target demand model to obtain the demand adaptation parameters.

[0008] In one embodiment, the step of obtaining new paradigm knowledge based on the knowledge object data, the verification data, and the applicable boundary data includes: determining whether the verification data and the applicable boundary data meet a first preset condition; when the verification data and the applicable boundary data meet the first preset condition, obtaining the extended knowledge based on the upgrade model and the knowledge object data, and obtaining the new paradigm knowledge based on the reshaping model, the knowledge object data, and the extended knowledge; when the verification data and the applicable boundary data do not meet the first preset condition, obtaining the new paradigm knowledge based on the reshaping model and the knowledge object data, wherein the extended knowledge is the result of optimizing knowledge within the same framework after adjusting parameters and optimizing processes of the knowledge object data, and the new paradigm knowledge is a new theoretical, methodological, and application system of knowledge formed by reconstructing cognition after breaking through the logical framework of the knowledge object data or the extended knowledge.

[0009] In one embodiment, determining whether the verification data and the applicable boundary data meet a first preset condition includes: weighting the consistency score, validity score, and adaptability score to obtain a comprehensive score; obtaining the applicable scope based on the applicable boundary data; and determining that the verification data and the applicable boundary data meet the first preset condition when the comprehensive score is greater than or equal to the preset score and the applicable scope is greater than or equal to the preset scope.

[0010] In one embodiment, when the verification data and the applicable boundary data satisfy a first preset condition, the extended knowledge is obtained based on the upgrade model and the knowledge object data, and the new paradigm knowledge is obtained based on the reshaping model, the knowledge object data, and the extended knowledge. This includes: performing multi-dimensional analysis on the knowledge object data through the upgrade model to obtain implicit constraint data; the upgrade model obtaining the extended knowledge based on the knowledge object data and the implicit constraint data; performing correlation reconstruction processing on the knowledge object data and the extended knowledge through the reshaping model to obtain reconstructed cognitive data; and the reshaping model obtaining the new paradigm knowledge based on the knowledge object data, the extended knowledge, and the reconstructed cognitive data.

[0011] In one embodiment, when the verification data and the applicable boundary data do not meet the first preset condition, the new paradigm knowledge is obtained based on the reshaping model and the knowledge object data, including: reconstructing the knowledge object data through the reshaping model to obtain reconstructed cognitive data; the reshaping model obtains the new paradigm knowledge based on the knowledge object data and the reconstructed cognitive data.

[0012] The present invention also provides a computing device, characterized in that it includes: a processor and a memory storing a computer program, wherein when the processor runs the computer program, it implements the steps of the above-described knowledge processing method.

[0013] The present invention also provides a storage medium storing a computer program, which, when executed by a processor, implements the steps of the above-described knowledge processing method.

[0014] This invention provides a knowledge processing method, computing device, and storage medium. Responding to user-sent information to be processed, it verifies the information using a verification model, acquires knowledge object data, verification data, and applicable boundary data, and obtains new paradigm knowledge based on the knowledge object data, verification data, and applicable boundary data. This invention can generate new paradigm knowledge based on information to be processed, improving the standardization and adaptability of knowledge processing. Attached Figure Description

[0015] Figure 1 This is a flowchart of the knowledge processing method according to the first embodiment of the present invention.

[0016] Figure 2 For the present invention Figure 1 The flowchart for step S11.

[0017] Figure 3 For the present invention Figure 1 The flowchart for step S12.

[0018] Figure 4 This is a schematic diagram of the structure of a knowledge processing system according to an embodiment of the present invention.

[0019] Figure 5 This is a schematic diagram of the structure of a computing device according to an embodiment of the present invention.

[0020] Figure reference numerals: Base layer - 41; Upgrade layer - 42; Remodeling layer - 43; Processor - 50; Memory - 51; Network interface - 52; Bus system - 53. Detailed Implementation

[0021] The foregoing and other technical contents, features, and effects of the present invention will be clearly presented in the following detailed description of preferred embodiments with reference to the accompanying drawings. Through the description of the specific embodiments, a more in-depth and specific understanding can be gained of the technical means and effects adopted by the present invention to achieve its intended purpose. However, the accompanying drawings are for reference and illustration only and are not intended to limit the present invention.

[0022] To further illustrate the technical means and effects adopted by the present invention in order to achieve the intended purpose, the present invention will be described in detail below with reference to the accompanying drawings and preferred embodiments.

[0023] Figure 1 This is a flowchart of the knowledge processing method according to the first embodiment of the present invention.

[0024] like Figure 1 As shown, the knowledge processing method provided in this embodiment includes the following steps: Step S11: In response to the pending information sent by the user, verify the pending information using the verification model, and obtain knowledge object data, verification data, and applicable boundary data.

[0025] like Figure 1 As shown, step S11 includes: Step S111: Perform semantic parsing on the information to be processed using the verification model to obtain knowledge object data.

[0026] Specifically, the information to be processed is input into the validation model, which performs semantic parsing to obtain knowledge object data. This knowledge object data includes conceptual information, logical relationship information, constraint information, application scenario information, and target indicator information. For example, the validation model performs semantic parsing on the statement: "This heat treatment process uses a salt bath furnace to quench 45# steel parts at a heating temperature of 820-840 degrees Celsius. The holding time is calculated based on the effective thickness of the part at 1.5 minutes / mm. The quenching medium is 10% brine. After cooling to room temperature, tempering is performed at a tempering temperature of 520 degrees Celsius for 2 hours. The final hardness requirement is HRC28-32. This process is suitable for mass production of small and medium-sized shaft parts, with the goal of minimizing energy consumption while ensuring hardness." The system identifies technical entities, such as salt bath furnace, No. 45 steel, quenching, tempering, 10% brine, and HRC, as conceptual information; relationships between entities, such as salt bath furnace → used for → quenching, and No. 45 steel → has → hardness requirements, as logical relationship information; numerical constraints, such as 820-840 degrees, 1.5 minutes / mm, and 520 degrees, as constraint information; mass production and small and medium-sized shaft parts as application scenario information; and ensuring hardness and reducing energy consumption as target indicator information.

[0027] Step S112: Perform system verification processing on the knowledge object data through the verification model to obtain verification data.

[0028] Specifically, the validation data includes consistency scores, validity scores, adaptability scores, conflict detection results, and matching calculation results. The validation model performs logical rule matching and conflict detection processing on conceptual information and logical relationship information to obtain conflict detection results and consistency scores. The validation model performs target matching calculation processing on target indicator information to obtain matching calculation results and validity scores. The validation model performs environmental simulation analysis processing on constraint information and application scenario information to obtain adaptability scores. For example, in logical rule matching and conflict detection processing, the validation model matches conceptual information (45# steel, 10% brine) and logical relationship information (quenching medium selection) with rules in the knowledge base. The knowledge base records that quenching 45# steel with 10% brine poses a cracking risk and recommends oil quenching or water quenching with oil cooling. The validation model detects this conflict, records the conflict detection result as a cracking risk in the quenching medium selection, and calculates the consistency score based on the severity of the conflict. In target matching calculation processing, the validation model matches target indicator information (ensuring hardness HRC28-32, reducing energy consumption) with the expected output of the process scheme. The model predicts that the hardness of 45 steel after tempering at 520 degrees Celsius will be approximately HRC28-32, with a hardness matching rate of 95%. The model calculates that the energy consumption per unit product of the salt bath furnace is approximately 0.8 kWh / kg, which is 40% higher than the industry-leading level of 0.5 kWh / kg. The verification model records the matching calculation results as meeting hardness standards but failing to meet energy consumption standards, and calculates an effectiveness score based on the weights of each objective. In environmental simulation analysis, the verification model conducts environmental simulation analysis based on constraint information (820-840 degrees Celsius, 1.5 minutes / mm, 10% brine) and application scenario information (mass production, small and medium-sized shaft parts). When simulating material fluctuation scenarios, the carbon content of 45 steel drops to the lower limit of 0.42%, and the simulation shows that the hardness after quenching is less than HRC28. When simulating environmental policy scenarios, stricter policies restrict the use of salt bath furnaces, rendering the solution ineffective. An adaptability score is obtained based on the environmental simulation analysis results of multiple scenarios.

[0029] Step S113: Perform extended verification processing on the knowledge object data through the verification model to obtain applicable boundary data.

[0030] Specifically, the applicable boundary data includes extensibility parameters and demand adaptation parameters. The validation model constructs module dependencies based on logical relationship information and performs structural analysis to obtain extensibility parameters. The validation model constructs a target demand model based on target indicator information and application scenario information, and performs matching calculations based on the application scenario information and the target demand model to obtain demand adaptation parameters. For example, in the structural analysis process, the validation model transforms logical relationship information into a directed graph structure, where nodes represent process modules (heating module, quenching module, tempering module, and detection module), and edges represent dependencies. Through graph theory algorithm analysis, the extensibility parameters are obtained based on the analysis results. In the matching calculation process, the validation model constructs a target demand vector based on target indicator information (ensuring hardness and reducing energy consumption) and application scenario information (mass production, small and medium-sized shaft parts), transforms the current process scheme into a capability vector, and obtains demand adaptation parameters by calculating cosine similarity.

[0031] Step S12: Based on the knowledge object data, the verification data, and the applicable boundary data, obtain new paradigm knowledge.

[0032] like Figure 3 As shown, step S12 includes: Step S121: Determine whether the verification data and the applicable boundary data meet the first preset condition.

[0033] Specifically, the consistency score, validity score, and adaptability score are weighted and calculated to obtain a comprehensive score. The applicable scope is determined based on the applicable boundary data. When the comprehensive score is greater than or equal to a preset score, and the applicable scope is greater than or equal to a preset range, the verification data and applicable boundary data are determined to meet a first preset condition. When the verification data and applicable boundary data meet the first preset condition, step S122 is entered: based on the upgrade model and the knowledge object data, the extended knowledge is obtained, and based on the reshaping model, the knowledge object data, and the extended knowledge, the new paradigm knowledge is obtained. When the verification data and applicable boundary data do not meet the first preset condition, step S123 is entered: based on the reshaping model and the knowledge object data, the new paradigm knowledge is obtained. Extended knowledge is the optimized result of knowledge within the same framework formed after parameter adjustment and process optimization of the knowledge object data; new paradigm knowledge is a completely new theoretical, methodological, and application system of knowledge formed by reconstructing cognition after breaking through the logical framework of the knowledge object data.

[0034] Specifically, step S122 includes upgrading the model to perform multidimensional analysis and processing on the knowledge object data to obtain implicit constraint data; upgrading the model to obtain extended knowledge based on the knowledge object data and implicit constraint data; reshaping the model to reconstruct the knowledge object data and extended knowledge to obtain reconstructed cognitive data; and reshaping the model to obtain new paradigm knowledge based on the knowledge object data, extended knowledge, and reconstructed cognitive data.

[0035] Implicit constraint data includes at least one of the following: institutional constraint data, sentiment bias data, cultural difference data, and language trap data. The upgraded model performs multi-dimensional analysis on knowledge object data, including: identifying the institutional context of logical relationship information to obtain institutional constraint data; performing sentiment semantic analysis on target indicator information to obtain sentiment bias data; performing cross-corpus comparison on application scenario information to obtain cultural difference data; and performing logical rule matching on logical relationship information to obtain at least one of the following:

[0036] The upgraded model, based on knowledge object data and implicit constraint data, acquires extended knowledge by: processing conflict detection results for conflict localization to obtain conflict correction data; filtering low-matching items from matching calculation results to obtain target optimization data; performing structural analysis on extended capability parameters to obtain structural optimization data; and performing constraint correction on implicit constraint data to obtain constraint correction data. Finally, based on conflict correction data, target optimization data, structural optimization data, and constraint correction data, the upgraded model performs parameter adjustment, process path optimization, and structural reorganization on knowledge object data to acquire extended knowledge. For example, the upgraded model addresses the risk of cracking in the quenching medium selection based on conflict detection results by locating the quenching medium parameters in the knowledge object data and obtaining conflict correction data suggesting changing the 10% brine solution to water quenching and oil cooling. The upgraded model also filters out low-matching items based on the energy consumption matching degree of 40% in the matching calculation results, identifying the tempering temperature of 520 degrees Celsius as a low-matching item and obtaining target optimization data suggesting reducing the tempering temperature to 450-480 degrees Celsius. Finally, the upgraded model performs structural analysis on the ductility parameters, identifying the strong coupling between quenching and tempering, and obtaining structural optimization data to decouple the quenching and tempering parameters, establishing an independent optimization space. The upgraded model performs constraint correction processing on implicit constraint data, acquiring constraint correction data to add environmental constraints. If environmental policies prohibit salt bath furnaces, the solution is switched to a vacuum furnace. Safety constraints are added, requiring anti-splash shields for quenching operations. Based on conflict correction data, target optimization data, structural optimization data, and constraint correction data, the upgraded model adjusts parameters of the knowledge object data (changing the quenching medium to water quenching and oil cooling, and adjusting the tempering temperature to 450-480 degrees Celsius), optimizes the process path (adding post-quenching hardness sampling inspection), and restructures the structure (decoupling heating, quenching, and tempering parameters into independent modules), acquiring optimized solution data as extended knowledge. This extended knowledge includes optimized heat treatment process schemes, newly added constraints, decoupled modular process structures, and improved quality control processes.

[0037] Then, the reshaping model performs difference analysis on the knowledge object data and extended knowledge to obtain incremental feature data. Incremental feature data is used to characterize the new, adjusted, or optimized content of the extended knowledge relative to the knowledge object data. For example, the reshaping model compares and analyzes the original heat treatment process with the extended knowledge, identifying changes such as the quenching medium being changed from 10% brine to water-quenched oil cooling, the tempering temperature being changed from 520 degrees to 450-480 degrees, the addition of a post-quenching hardness sampling inspection process, and new environmental constraints. These changes are then used as incremental feature data.

[0038] The remodeling model performs source analysis on knowledge object data and incremental feature data to obtain extended underlying hypothesis data. For example, the remodeling model analyzes the reason for changing the quenching medium from 10% brine to water quenching and oil cooling, and determines that its underlying hypothesis is that reducing the risk of cracking is more important than increasing the cooling rate; it analyzes the reason for adding a hardness sampling inspection process, and determines that its underlying hypothesis is that process verification can improve quality stability; it analyzes the reason for adding environmental constraints, and determines that its underlying hypothesis is that process sustainability has become an important goal.

[0039] The reshaping model performs cross-domain mapping processing on the extended underlying assumption data to obtain extended cross-domain mapping data. For example, the reshaping model maps new environmental constraints to resource cycling mechanisms in ecosystems to obtain mapping knowledge of adapting to changes in the external environment; maps new quality sampling processes to feedback regulation mechanisms in cybernetics to obtain mapping knowledge of closed-loop process control; and maps module decoupling structures to microservice architecture in software engineering to obtain mapping knowledge of functional independence and collaborative optimization.

[0040] The reshaping model performs essential abstraction on the extended underlying assumption data and extended cross-domain mapping data to obtain extended core feature data. For example, the reshaping model abstracts the core feature that the system needs to continuously adapt to changes in external rules from the mapping of environmental constraints and ecosystems; it abstracts the core feature that process self-correction is better than result correction from the mapping of quality sampling and feedback control; and it abstracts the core feature that local optimization and overall collaboration coexist from the mapping of module decoupling and microservice architecture.

[0041] The reshaping model combines and reconstructs data based on constraint information in knowledge object data, newly added constraints in extended knowledge, and expanded core feature data. It then filters this data based on target indicator information to obtain reconstructed cognitive data. For example, the reshaping model reorganizes core features such as dynamic adaptation to changes in external rules, process self-correction, and module collaborative optimization to form a new cognitive framework: heat treatment process is not only a parameter control process but also an adaptive control process that continuously evolves in response to targets, environment, and feedback information. The reconstructed cognitive data includes adaptive process cognition, dynamic feedback cognition, and collaborative optimization cognition.

[0042] The reshaping model, based on knowledge object data, extended knowledge, and reconstructed cognitive data, acquires new paradigm knowledge by fusing the original logical relationships in the knowledge object data, the optimized logical relationships in the extended knowledge, and the new cognitive relationships in the reconstructed cognitive data to obtain theoretical model data. For example, the original theory emphasizes controlling product hardness with fixed process parameters, the extended knowledge emphasizes optimizing parameters and process flow, and the reconstructed cognitive data emphasizes dynamic feedback and environmental adaptability. The fusion of these three forms "theoretical model data for adaptive heat treatment," which achieves product performance control through real-time perception, dynamic decision-making, and continuous optimization.

[0043] The remodeling model generates corresponding methodological process data based on theoretical model data. For example, the original fixed-parameter execution process is reconstructed into a closed-loop process of prediction, execution, monitoring, feedback, and optimization, and environmental constraint changes, quality inspection results, and energy consumption indicators are incorporated into the decision-making process to achieve dynamic adjustment of the process.

[0044] The model reshaping generates application system structure data based on methodological process data. For example, it constructs a closed-loop control system structure including a perception module, a prediction module, a decision-making module, an execution module, and a learning and optimization module, where the learning and optimization module continuously updates the decision-making strategy based on historical operating data.

[0045] The remodeling model integrates and processes theoretical model data, methodological process data, and application system structure data to obtain new paradigm knowledge. For example, the new paradigm knowledge no longer takes fixed heat treatment parameters as its core, but forms an intelligent heat treatment system with real-time perception, dynamic feedback, adaptive optimization, and continuous evolution as its core, thereby achieving a breakthrough in the original knowledge framework.

[0046] Specifically, step S123 includes reshaping the knowledge object data by the reshaping model to obtain reconstructed cognitive data, and then using the knowledge object data and the reconstructed cognitive data to obtain new paradigm knowledge.

[0047] The reshaping model reconstructs knowledge object data to obtain reconstructed cognitive data, including: The reshaping model performs source analysis on knowledge object data to obtain underlying hypothesis data. For example, the reshaping model uses recursive questioning techniques to perform source analysis on heat treatment process schemes. The first question is: Why use a salt bath furnace? Because salt bath furnaces provide uniform heating and precise temperature control. The second question is: Why is uniform heating and precise temperature control necessary? Because it's necessary to ensure consistent hardness across the parts. The third question is: Why ensure consistent hardness? Because the parts need to withstand uniform loads during use. The fourth question is: Why is it necessary to withstand uniform loads? Because parts will wear out, and consistent hardness extends their lifespan. The fifth question is: Why extend lifespan? Because it's necessary to reduce part replacement costs. Through recursive questioning, the reshaping model can extract the underlying hypothesis data: uniform heating is a prerequisite for ensuring consistent hardness; hardness is a key indicator determining part lifespan; and extending part lifespan is the core objective.

[0048] The reshaping model performs cross-domain mapping processing on knowledge object data to obtain cross-domain mapping data. For example, based on the analogical reasoning algorithm of knowledge graphs, the reshaping model performs cross-domain mapping. With biology, it maps heat treatment processes to the metabolic processes of organisms, with heating corresponding to energy intake, quenching to stress response, and tempering to repair and regulation. The extracted cross-domain mapping data shows that organisms adapt to environmental changes through dynamic adjustment, rather than fixed parameters. With cybernetics, it maps open-loop control processes to closed-loop feedback systems. The extracted cross-domain mapping data shows that dynamic adjustment based on real-time feedback is more effective than fixed parameter control. With economics, it maps process parameter optimization to resource optimization allocation. The extracted cross-domain mapping data finds the optimal solution under multiple constraints, rather than a single objective optimization.

[0049] The reshaping model performs essential abstraction processing on the underlying assumption data and cross-domain mapping data to obtain core feature data. For example, the reshaping model uses a conceptual abstraction algorithm for essential extraction. From the assumption that uniform heating is a prerequisite for ensuring consistent hardness and the biological mapping that organisms adapt to the environment through dynamic adjustment, it abstracts the core feature that uniformity is not the only path and that dynamic adjustment can also achieve consistency. From the assumption that hardness is a key indicator determining the lifespan of a part and the economic mapping that it seeks the optimal solution under multiple constraints, it abstracts the core feature that hardness is only a characterization indicator of lifespan, and that microstructure is the essential determining factor. From the assumption that extending the lifespan of a part is the core objective and the cybernetics mapping that closed-loop feedback is more effective, it abstracts the core feature that the focus shifts from goal-driven to process-driven, and that real-time feedback adjustment is superior to preset parameters. The core feature data is recorded as dynamic adjustment being superior to uniformity, microstructure determining performance, and process feedback replacing preset parameters.

[0050] The reshaping model extracts and formalizes logical relationship and constraint information to obtain structural constraints. For example, the reshaping model extracts all constraint information from the knowledge object data and converts it into formal logical expressions. The constraint heating temperature of 820-840 degrees is formalized as T_heating∈[820,840], the constraint holding time calculated at 1.5 minutes / mm is formalized as t_hold=1.5×D_effective, the constraint quenching medium of 10% brine is formalized as medium="brine", and the constraint final hardness HRC28-32 is formalized as hardness∈[28,32]. The above constraints are combined into structural constraints: S={T_heating∈[820,840], t_hold=1.5×D, medium="brine", hardness∈[28,32]}.

[0051] The reshaping model performs target modeling on target index information and application scenario information to obtain optimization objectives. For example, the reshaping model transforms target index information and application scenario information into a multi-objective optimization problem. The target index information ensures hardness and is formalized as min|hardness-target_hardness|, the target index information reduces energy consumption and is formalized as minenergy_consumption, the application scenario information adds the objective min production_time for batch production, and the application scenario information for small and medium-sized shaft parts adds the constraint D_shaft≤50mm. The optimization objective is recorded as O={min|H-30|, min E, minT, st D≤50}.

[0052] The reshaping model constructs a feature structure model based on core feature data and performs combined reconstruction processing on the feature structure model based on preset reconstruction rules. During the combined reconstruction process, the reshaping model performs legality verification based on structural constraints and filters the reconstruction results based on optimization objectives to obtain reconstructed cognitive data. For example, the reshaping model constructs a feature structure model based on core feature data, with micro-organizational control as the core node, real-time feedback adjustment as the connection mechanism, and dynamic adaptation as the behavioral characteristic. It then performs combined reconstruction processing to generate multiple reconstruction schemes: Scheme A constructs an organization-performance prediction model based on micro-organizational control; Scheme B constructs an online monitoring-dynamic control system based on real-time feedback adjustment; Scheme C combines A and B to construct a closed-loop intelligent heat treatment system. Legality verification shows that Scheme A does not violate the original physical constraints and passes; Scheme B requires the addition of sensor equipment but does not violate safety constraints and passes; Scheme C requires reconstructing the equipment layout, which may violate factory space constraints and requires further evaluation. Based on the optimized target selection, Scheme A has a hardness prediction error of ±2HRC and a 10% reduction in energy consumption, with a comprehensive score of 75; Scheme B has a hardness fluctuation of ±1HRC and a 5% reduction in energy consumption, with a comprehensive score of 80; Scheme C has a hardness fluctuation of ±0.5HRC and a 20% reduction in energy consumption, with a comprehensive score of 92. Scheme C, with the highest comprehensive score, was selected. Its reconstructed cognitive data includes: a new cognitive framework, recognizing that heat treatment is essentially about microstructure control rather than hardness control; a new control mode based on closed-loop feedback control with real-time monitoring, rather than open-loop parameter control; and a new evaluation system with microstructure uniformity as the core indicator, supplemented by multi-dimensional evaluations such as hardness and toughness.

[0053] The reshaping model, based on knowledge object data and reconstructed cognitive data, acquires new paradigm knowledge, including: The reshaping model processes reconstructed cognitive data through conceptual abstraction and relational reconstruction to obtain theoretical model data. For example, the reshaping model constructs a completely new theoretical framework based on reconstructed cognitive data. In terms of conceptual abstraction, new concepts are extracted from the reconstructed cognition; microstructure control replaces hardness control as the core objective; real-time feedback adjustment replaces preset parameter execution as the control method; and dynamic adaptation replaces fixed processes as the system characteristic. In terms of relational reconstruction, the original relationship of heating temperature → hardness is reconstructed as temperature field distribution → phase transformation kinetics → microstructure → macroscopic properties; and the original relationship of process parameters → product quality is reconstructed as real-time monitoring data → dynamic adjustment commands → process stability → product quality consistency. The theoretical model data corresponds to the essence of heat treatment being the control of the microstructure evolution process of materials, achieving precise control of product performance through real-time monitoring and dynamic feedback adjustment.

[0054] The remodeling model combines theoretical model data with rules and generates process flow data to obtain methodological flow data. For example, the remodeling model generates new methodologies based on theoretical model data. In terms of rule combination, it generates rules, establishes a heat treatment process model to simulate the temperature field and microstructure evolution in real time, deploys multi-point temperature sensors and online microstructure detection devices to acquire real-time data, inputs the real-time data into the model to predict the microstructure evolution trend in the next 5 minutes, dynamically adjusts the heating power and cooling medium flow rate based on the prediction results, and records the data of each processing to continuously optimize the model parameters. In terms of process generation, the generation steps are: 1. Load and number the parts, read the part specification parameters; 2. Start the model to pre-simulate the optimal process path; 3. Execute heat treatment; 4. Collect temperature and microstructure data in real time; 5. The model compares the actual data with the predicted data in real time. When the deviation exceeds the threshold, it triggers adjustment, adjusting the heating power or cooling medium flow rate to make the actual curve approach the optimal path; 6. After processing, record the entire process data and update the model parameters.

[0055] The remodeling model performs module mapping and system structure configuration processing on the method flow data to obtain application system structure data. For example, the remodeling model maps method flow data to an implementable technical system. In terms of module mapping, method steps 1-2 are mapped to an intelligent planning module (including a parts database, process simulation engine, and path optimization algorithm); method steps 3-4 are mapped to a sensing module (including a multi-point thermocouple array, online tissue detection sensor, and data acquisition unit); method step 5 is mapped to a control module (including a PID controller, variable frequency heating power supply, and adjustable flow cooling system); and method step 6 is mapped to a learning module (including a data recording unit, model training engine, and parameter update algorithm). In terms of system structure configuration, the sensing module collects data in real time and transmits it to the control module via industrial Ethernet; the control module synchronizes the data to the model of the intelligent planning module, the intelligent planning module runs simulation prediction, and issues adjustment commands to the control module, which executes the commands to adjust the actuators; the learning module processes historical data offline and updates model parameters periodically.

[0056] The model reshaping process integrates and processes the theoretical model data, methodological process data, and application system structure data to obtain new paradigm data. This new paradigm data is used to characterize the new theoretical, methodological, and application systems formed after breaking through the original knowledge framework.

[0057] Unlike upgrade models, which optimize and expand existing knowledge, reshaping models reconstruct knowledge object data through critical and creative thinking mechanisms. Critical thinking identifies implicit assumptions, logical inertia, goal biases, and structural limitations in the knowledge object data and analyzes their rationality. Creative thinking generates new cognitive structures through cross-domain mapping, concept transfer, essential abstraction, and relational restructuring, thereby forming new paradigmatic knowledge that breaks through the original knowledge framework.

[0058] In one implementation, the verification model, upgrade model, and remodel model are invoked via a network proxy service.

[0059] In one implementation, after receiving the pending information sent by the user, the system sends the user the extended knowledge and new paradigm knowledge corresponding to the pending information.

[0060] This invention provides a knowledge processing method that, in response to user-sent information to be processed, verifies the information using a verification model, acquires knowledge object data, verification data, and applicable boundary data, and obtains new paradigm knowledge based on the knowledge object data, verification data, and applicable boundary data. This invention can generate extended knowledge or new paradigm knowledge based on the information to be processed, improving the standardization and adaptability of knowledge processing.

[0061] Figure 4This is a schematic diagram of the structure of a knowledge processing system according to an embodiment of the present invention.

[0062] like Figure 4 As shown, the knowledge processing system includes a basic layer 41, an upgrade layer 42, and a reshaping layer 43.

[0063] The base layer 41 is used to respond to the pending information sent by the user, verify the pending information using a verification model, obtain knowledge object data, verification data, and applicable boundary data, determine whether the verification data and applicable boundary data meet the first preset condition, and send the knowledge object data to the upgrade layer 42 when the verification data and applicable boundary data meet the first preset condition, and send the knowledge object data to the reshaping layer 43 when the verification data and applicable boundary data do not meet the first preset condition.

[0064] The foundation layer 41 is used to: perform semantic parsing processing on the information to be processed through the validation model to obtain knowledge object data; perform system validation processing on the knowledge object data through the validation model to obtain validation data; and perform extended validation processing on the knowledge object data through the validation model to obtain applicable boundary data. Knowledge object data includes conceptual information, logical relationship information, target indicator information, constraint information, and application scenario information. Validation data includes consistency scores, effectiveness scores, and adaptability scores. Applicable boundary data includes extended capability parameters and requirement adaptation parameters.

[0065] The basic layer 41 is used to perform logical rule matching and conflict detection processing on conceptual information and logical relationship information based on the verification model to obtain a consistency score; to perform target matching calculation processing on target indicator information based on the verification model to obtain an effectiveness score; and to perform environmental simulation analysis processing on constraint information and application scenario information based on the verification model to obtain an adaptability score.

[0066] The base layer 41 is used to construct module dependencies based on the logical relationship information of the verification model and perform structural analysis to obtain extensibility parameters; to construct a target requirement model based on the target indicator information and application scenario information of the verification model, and to perform matching calculations between the application scenario information and the target requirement model to obtain requirement adaptation parameters.

[0067] The base layer 41 is used to calculate a comprehensive score by weighting the consistency score, validity score and adaptability score; to obtain the applicable scope based on the applicable boundary data; and to determine that the verification data and applicable boundary data meet the first preset condition when the comprehensive score is greater than or equal to the preset score and the applicable scope is greater than or equal to the preset scope.

[0068] Upgrade layer 42 is used to obtain extended knowledge based on knowledge object data. Extended knowledge is the optimized knowledge result of the same framework formed after parameter adjustment and process optimization of knowledge object data.

[0069] Upgrade layer 42 is used to perform multidimensional analysis and processing on knowledge object data through the upgrade model to obtain implicit constraint data; and to process knowledge object data and implicit constraint data through the upgrade model to obtain extended knowledge.

[0070] The reshaping layer 43 is used to acquire new paradigm knowledge based on knowledge object data and / or extended knowledge. New paradigm knowledge is a new theoretical, methodological, and application system of knowledge formed by reconstructing cognition after breaking through the logical framework of knowledge object data.

[0071] The reshaping layer 43 is used to reconstruct knowledge object data and / or extended knowledge through the reshaping model to obtain reconstructed cognitive data; and to process knowledge object data and reconstructed cognitive data through the reshaping model to obtain new paradigm knowledge.

[0072] The knowledge processing system in this embodiment is used to implement the knowledge processing method described above. The functions implemented by each module of the knowledge processing system can be found in [reference needed]. Figures 1-3 The descriptions in the embodiments are not repeated here.

[0073] The present invention also provides a storage medium storing a computer program, which, when executed by a processor, implements the steps of the knowledge processing method described above.

[0074] This invention provides a knowledge processing method, computing device, and storage medium. Responding to user-sent information to be processed, it verifies the information using a verification model, acquires knowledge object data, verification data, and applicable boundary data, and obtains new paradigm knowledge based on the knowledge object data, verification data, and applicable boundary data. This invention can generate extended knowledge or new paradigm knowledge based on the information to be processed, improving the standardization and adaptability of knowledge processing.

[0075] like Figure 5 As shown, the present invention also provides a computing device, which includes: a processor 50 and a memory 51 storing a computer program; wherein, Figure 5 The processor 50 shown in the diagram does not indicate that there is only one processor 50, but rather indicates the positional relationship of the processor 50 relative to other devices. In practical applications, there can be one or more processors 50; similarly, Figure 5 The memory 51 shown in the diagram has the same meaning, that is, it is only used to indicate the positional relationship of memory 51 relative to other devices. In practical applications, there can be one or more memories 51. When the processor 50 runs the computer program, it implements the above-mentioned knowledge processing method.

[0076] The computing device may also include at least one network interface 52. Various components within the computing device are coupled together via a bus system 53. It is understood that the bus system 53 is used to implement communication between these components. In addition to a data bus, the bus system 53 also includes a power bus, a control bus, and a status signal bus. However, for clarity, in… Figure 5 The general labeled all buses as Bus System 53.

[0077] The memory 51 can be volatile memory or non-volatile memory, or both. Non-volatile memory can be read-only memory (ROM), programmable read-only memory (PROM), erasable programmable read-only memory (EPROM), electrically erasable programmable read-only memory (EEPROM), ferromagnetic random access memory (FRAM), flash memory, magnetic surface memory, optical disc, or compact disc read-only memory (CD-ROM); magnetic surface memory can be disk storage or magnetic tape storage. Volatile memory can be random access memory (RAM), used as an external cache. By way of example, but not limitation, many forms of RAM are available, such as Static Random Access Memory (SRAM), Synchronous Static Random Access Memory (SSRAM), Dynamic Random Access Memory (DRAM), Synchronous Dynamic Random Access Memory (SDRAM), Double Data Rate Synchronous Dynamic Random Access Memory (DDRSDRAM), Enhanced Synchronous Dynamic Random Access Memory (ESDRAM), SyncLink Dynamic Random Access Memory (SLDRAM), and Direct Rambus Random Access Memory (DRRAM).The memory 51 described in the embodiments of the present invention is intended to include, but is not limited to, these and any other suitable types of memory.

[0078] The memory 51 in this embodiment of the invention is used to store various types of data to support the operation of the computing device. Examples of this data include: any computer programs used to operate on the computing device, such as operating systems and applications; contact data; phonebook data; messages; pictures; videos, etc. The operating system includes various system programs, such as the framework layer, core library layer, driver layer, etc., used to implement various basic services and handle hardware-based tasks. Applications can include various applications, such as media players, browsers, etc., used to implement various application services. Here, the program implementing the knowledge processing method of this embodiment of the invention can be included in the application.

[0079] The above are merely preferred embodiments of the present invention and are not intended to limit the present invention in any way. Although the present invention has been disclosed above with reference to preferred embodiments, it is not intended to limit the present invention. Any person skilled in the art can make some modifications or alterations to the above-disclosed technical content to create equivalent embodiments without departing from the scope of the present invention. Any simple modifications, equivalent changes and alterations made to the above embodiments based on the technical essence of the present invention without departing from the scope of the present invention shall still fall within the scope of the present invention.

Claims

1. A knowledge processing method, characterized in that, The method includes: In response to the pending information sent by the user, the pending information is verified using a verification model, and knowledge object data, verification data, and applicable boundary data are obtained. Based on the knowledge object data, the verification data, and the applicable boundary data, new paradigm knowledge is obtained.

2. The knowledge processing method as described in claim 1, characterized in that, The steps of responding to user-sent pending information, verifying the pending information using a verification model, and obtaining knowledge object data, verification data, and applicable boundary data include: The verification model is used to perform semantic parsing on the information to be processed to obtain knowledge object data. The knowledge object data is systematically verified using the verification model to obtain verification data. The knowledge object data is extended and validated using the validation model to obtain applicable boundary data.

3. The knowledge processing method as described in claim 2, characterized in that, The knowledge object data includes conceptual information, logical relationship information, target indicator information, constraint information, and application scenario information; the verification data includes consistency score, validity score, and adaptability score. The knowledge object data is systematically verified using the verification model to obtain verification data, including: The verification model performs logical rule matching and conflict detection processing based on the conceptual information and the logical relationship information to obtain the consistency score; The verification model performs target matching calculation based on the target indicator information to obtain the validity score; The verification model performs environmental simulation analysis based on the constraint information and the application scenario information to obtain the adaptability score.

4. The knowledge processing method as described in claim 3, characterized in that, Applicable boundary data includes scalability parameters and demand adaptation parameters; The knowledge object data is extended and validated using the validation model to obtain applicable boundary data, including: The verification model constructs module dependencies based on the logical relationship information and performs structural analysis to obtain the extensibility parameters. The verification model constructs a target requirement model based on the target indicator information and the application scenario information, and performs matching calculations between the application scenario information and the target requirement model to obtain the requirement adaptation parameters.

5. The knowledge processing method as described in claim 1, characterized in that, The step of obtaining new paradigm knowledge based on the knowledge object data, the verification data, and the applicable boundary data includes: Determine whether the verification data and the applicable boundary data satisfy a first preset condition; When the verification data and the applicable boundary data meet the first preset condition, the extended knowledge is obtained based on the upgrade model and the knowledge object data, and the new paradigm knowledge is obtained based on the reshaping model, the knowledge object data and the extended knowledge; When the verification data and the applicable boundary data do not meet the first preset condition, the new paradigm knowledge is obtained based on the reshaping model and the knowledge object data; the extended knowledge is the same framework knowledge optimization result formed after parameter adjustment and process optimization of the knowledge object data; the new paradigm knowledge is a new theoretical, methodological, and application system knowledge formed by reconstructing cognition after breaking through the logical framework of the knowledge object data or the extended knowledge.

6. The knowledge processing method as described in claim 5, characterized in that, Determining whether the verification data and the applicable boundary data satisfy the first preset condition includes: The consistency score, validity score, and adaptability score are weighted and calculated to obtain a comprehensive score. The applicable scope is obtained based on the applicable boundary data; When the overall score is greater than or equal to the preset score, and the applicable scope is greater than or equal to the preset range, the verification data and the applicable boundary data are determined to satisfy the first preset condition.

7. The knowledge processing method as described in claim 5, characterized in that, When the verification data and the applicable boundary data meet the first preset condition, based on the upgrade model and the knowledge object data, the extended knowledge is obtained, and based on the reshaping model, the knowledge object data, and the extended knowledge, the new paradigm knowledge is obtained, including: The knowledge object data is analyzed and processed in multiple dimensions using the upgrade model to obtain implicit constraint data. The upgrade model obtains the extended knowledge based on the knowledge object data and the implicit constraint data; The knowledge object data and the extended knowledge are correlated and reconstructed using the reshaping model to obtain reconstructed cognitive data. The reshaping model obtains the new paradigm knowledge based on the knowledge object data, the extended knowledge, and the reconstructed cognitive data.

8. The knowledge processing method as described in claim 5, characterized in that, When the verification data and the applicable boundary data do not meet the first preset condition, the new paradigm knowledge is obtained based on the reshaping model and the knowledge object data, including: The knowledge object data is reconstructed using the reshaping model to obtain reconstructed cognitive data. The reshaping model obtains the new paradigm knowledge based on the knowledge object data and the reconstructed cognitive data.

9. A computing device, characterized in that, include: A processor and a memory storing a computer program, wherein, when the processor executes the computer program, a method for processing the knowledge according to any one of claims 1 to 8 is implemented.

10. A storage medium, characterized in that, The storage medium stores computer execution instructions, which, when executed by a processor, are used to implement the steps of the knowledge processing method as described in any one of claims 1 to 8.

Images