A large model reasoning method and related equipment for stem cell production and preparation

Abstract

The present application relates to the field of sensor data analysis technology, and provides a large model reasoning method and related equipment for stem cell production and preparation, the method comprising: obtaining multiple sensor data at the current moment in the stem cell production and preparation Internet of Things; constructing a first prompt template based on all sensor data at the current moment, and using the large model to reason on the first prompt template to obtain a first prediction result; when all sensor data at the current moment and the first prediction result meet the reasoning mode switching condition, constructing a second prompt template based on all sensor data at the current moment and the output format of the large model, and using the large model to reason on the second prompt template to obtain a second prediction result. The method of the present application can improve the timeliness of abnormality detection of the stem cell production and preparation environment using large model reasoning.

Description

Technical Field

[0001] The present application relates to the field of sensor data analysis technology, and in particular to a large model reasoning method and related equipment for stem cell production and preparation. Background Art

[0002] Monitoring and analyzing sensor data is of great significance in the production and preparation environment of stem cells. Most of the research on machine learning in sensor systems focuses on low-level perception tasks that process raw sensor data in a short period of time, while many practical applications require more advanced reasoning capabilities to understand and reason concepts from long-term sensor data. Existing machine learning-based methods have poor generalization capabilities when data is limited. Therefore, existing research has proposed a concept of pervasive artificial intelligence, that is, large models can interact and reason with the physical world through IoT sensors. Large models have strong reasoning capabilities and rich world knowledge, which can be used to interpret sensor data and perform high-level reasoning, and identify complex events in long-term sensor data. High-level reasoning tasks usually require strong reasoning capabilities to interpret complex sensor data and integrate domain knowledge to make decisions.

[0003] In order to effectively reason about the sensor data of the stem cell production and preparation environment and realize the interaction and reasoning of the large model with the physical world, the researchers used static prompts to reason about the sensor signal data. The prompt template was frozen once it was set. Only new sensor data needed to be provided for new reasoning without further changing the prompt. The singleness of the static prompt undoubtedly limited the ability of the large model in complex tasks. Then, in order to further improve the ability of the large model to handle complex tasks, the researchers proposed to use machine learning models to complete low-level perception tasks of sensor data, and then input the perception results into the large model to assist in high-level reasoning. However, this method has requirements for the performance of the machine learning model, and the single output of the perception result will miss the potential information in the data, affecting the reasoning performance of the large model in complex tasks. It can be seen that the current large model reasoning method has the problem of poor timeliness in using large model reasoning to detect anomalies in the stem cell production and preparation environment. Summary of the invention

[0004] The present application provides a large-model reasoning method and related equipment for stem cell production and preparation, which can solve the problem of poor timeliness of abnormality detection in the stem cell production and preparation environment using large-model reasoning.

[0005] In a first aspect, the present application provides a large model reasoning method for stem cell production and preparation, the large model reasoning method comprising: Obtain multiple sensor data at the current moment in the stem cell production and preparation Internet of Things; A first prompt template is constructed based on all sensor data at the current moment, and the first prompt template is inferred using the large model to obtain a first prediction result; the first prediction result is used to describe whether the detection index corresponding to each sensor at the current moment is abnormal, and the first prompt template is used to prompt the large model to perform abnormal detection on the detection index; When all sensor data at the current moment and the first prediction result meet the inference mode switching condition, a second prompt template is constructed based on all sensor data at the current moment and the output format of the large model, and the large model is used to infer the second prompt template to obtain a second prediction result; the second prediction result is used to describe the abnormal conditions of the detection indicators corresponding to each sensor at the future moment, and the second prompt template is used to prompt the large model to predict the abnormal conditions of the detection indicators.

[0006] Optionally, a first prompt template is constructed based on all sensor data at the current moment, including: By formula: Constructing the First Prompt Template ; in, represents the first target of the large model, represents the mode threshold, Represents all sensor data at the current moment, Represents the first domain knowledge.

[0007] Optionally, the inference mode switching condition is: The first prediction result describes that at the current moment, there is at least one sensor data corresponding to an abnormal detection indicator, or at least one sensor data among all the sensor data at the current moment has a value that is not within the numerical range described by the mode threshold.

[0008] Optionally, a second prompt template is constructed based on all sensor data at the current moment and the output format of the large model, including: By formula: Constructing the Second Prompt Template ; in, represents the second goal of the large model, represents the mode threshold, Represents all sensor data at the current moment, represents the second domain knowledge, Represents the output format for large models.

[0009] Optionally, the large model reasoning method also includes: When the end moment of the reasoning cycle arrives, the sensor data of all sensors at each moment in the reasoning cycle are used to generate examples for each moment, a third prompt template is constructed based on all examples, and the third prompt template is inferred using the large model to obtain long-term reasoning prediction results; the long-term reasoning prediction results are used to describe the data changes of the detection indicators within the reasoning cycle.

[0010] Optionally, using sensor data from all sensors at each moment in the inference cycle, generate examples for each moment, including: By formula: Generate Examples of moments ; in, Indicates the problem target, Indicates The triplet corresponding to each moment is , Indicates a moment, Indicates All sensor data at a moment in time, Indicates The actual result of the moment, , represents the reasoning chain, To express an answer; Build a third prompt template based on all examples, including: By formula: Constructing the Third Prompt Template ; in, Indicates the third goal, represents the mode threshold, represents the set of triples corresponding to all moments, Represents the third domain knowledge, Represents the set of examples corresponding to all moments.

[0011] Optionally, after the step of using the large model to infer the second prompt template to obtain the second prediction result, the large model inference method further includes: By formula: Update the mode threshold to obtain the updated mode threshold ; in, Indicates the lower limit of the numerical range described by the mode threshold. Indicates the upper limit of the numerical range described by the mode threshold. Represents all sensor data at the current moment.

[0012] In a second aspect, the present application provides a large model reasoning device for stem cell production and preparation, comprising: An acquisition module is used to acquire multiple sensor data at the current moment in the stem cell production and preparation Internet of Things; A first inference module is used to construct a first prompt template based on all sensor data at the current moment, and use the large model to infer the first prompt template to obtain a first prediction result; the first prediction result is used to describe whether the detection index corresponding to each sensor at the current moment is abnormal, and the first prompt template is used to prompt the large model to perform abnormal detection on the detection index; The second reasoning module is used to construct a second prompt template based on all sensor data at the current moment and the output format of the large model when all sensor data at the current moment and the first prediction result meet the reasoning mode switching condition, and use the large model to infer the second prompt template to obtain a second prediction result; the second prediction result is used to describe the abnormal condition of the detection indicator corresponding to each sensor at a future moment, and the second prompt template is used to prompt the large model to predict the abnormal condition of the detection indicator.

[0013] In a third aspect, an embodiment of the present application provides a terminal device, including a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor implements the above-mentioned large-model reasoning method for stem cell production and preparation when executing the above-mentioned computer program.

[0014] In a fourth aspect, an embodiment of the present application provides a computer-readable storage medium, which stores a computer program, and when the computer program is executed by a processor, it implements the above-mentioned large-model reasoning method for stem cell production and preparation.

[0015] The above solution of the present application has the following beneficial effects: In an embodiment of the present application, by acquiring sensor data of multiple sensors at the current moment in the stem cell production and preparation Internet of Things, a first prompt template is constructed based on all sensor data at the current moment, and the first prompt template is inferred using a large model to obtain a first prediction result. When all sensor data at the current moment and the first prediction result meet the inference mode switching condition, a second prompt template is constructed based on all sensor data at the current moment and the output format of the large model, and the second prompt template is inferred using the large model to obtain a second prediction result. Among them, by adaptively switching the large model reasoning process, the large model is automatically able to reason about sensor data in short-term and abnormal conditions, meeting different reasoning and prediction requirements, and adaptively monitoring and predicting the stem cell production and preparation environment in real time, ensuring the timeliness and reasoning depth of the large model reasoning, and effectively improving the timeliness of abnormal detection of the stem cell production and preparation environment using large model reasoning.

[0016] Other beneficial effects of the present application will be described in detail in the subsequent specific implementation section. BRIEF DESCRIPTION OF THE DRAWINGS

[0017] In order to more clearly illustrate the technical solutions in the embodiments of the present application, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present application. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying any creative work.

[0018] Figure 1 A flowchart of a large model reasoning method for stem cell production and preparation provided in one embodiment of the present application; Figure 2 A flowchart of a sentinel mode provided in an embodiment of the present application; Figure 3 A flowchart of an early warning mode provided in an embodiment of the present application; Figure 4 A flow chart of a detection mode provided in an embodiment of the present application; Figure 5 A schematic diagram of the structure of a large-scale model reasoning device for stem cell production and preparation provided in one embodiment of the present application; Figure 6 A schematic diagram of the structure of a terminal device provided in one embodiment of the present application. DETAILED DESCRIPTION

[0019] In the following description, specific details such as specific system structures, technologies, etc. are provided for the purpose of illustration rather than limitation, so as to provide a thorough understanding of the embodiments of the present application. However, it should be clear to those skilled in the art that the present application may also be implemented in other embodiments without these specific details. In other cases, detailed descriptions of well-known systems, devices, circuits, and methods are omitted to prevent unnecessary details from obstructing the description of the present application.

[0020] It should be understood that when used in the present specification and the appended claims, the term "comprising" indicates the presence of described features, wholes, steps, operations, elements and / or components, but does not exclude the presence or addition of one or more other features, wholes, steps, operations, elements, components and / or combinations thereof.

[0021] It should also be understood that the term “and / or” used in the specification and appended claims refers to any and all possible combinations of one or more of the associated listed items, and includes these combinations.

[0022] As used in the specification and appended claims of this application, the term "if" can be interpreted as "when" or "uponce" or "in response to determining" or "in response to detecting", depending on the context. Similarly, the phrase "if it is determined" or "if [described condition or event] is detected" can be interpreted as meaning "uponce it is determined" or "in response to determining" or "uponce [described condition or event] is detected" or "in response to detecting [described condition or event]", depending on the context.

[0023] In addition, in the description of the present application specification and the appended claims, the terms "first", "second", "third", etc. are only used to distinguish the descriptions and cannot be understood as indicating or implying relative importance.

[0024] References to "one embodiment" or "some embodiments" etc. described in the specification of this application mean that one or more embodiments of the present application include specific features, structures or characteristics described in conjunction with the embodiment. Therefore, the statements "in one embodiment", "in some embodiments", "in some other embodiments", "in some other embodiments", etc. that appear in different places in this specification do not necessarily refer to the same embodiment, but mean "one or more but not all embodiments", unless otherwise specifically emphasized in other ways. The terms "including", "comprising", "having" and their variations all mean "including but not limited to", unless otherwise specifically emphasized in other ways.

[0025] In view of the problem of poor timeliness of anomaly detection in the production and preparation environment of stem cells using large model reasoning, an embodiment of the present application provides a large model reasoning method for stem cell production and preparation. The large model reasoning method obtains sensor data of multiple sensors at the current moment in the stem cell production and preparation Internet of Things, then builds a first prompt template based on all sensor data at the current moment, and uses the large model to reason on the first prompt template to obtain a first prediction result. When all sensor data at the current moment and the first prediction result meet the reasoning mode switching condition, a second prompt template is built based on all sensor data at the current moment and the output format of the large model, and the large model is used to reason on the second prompt template to obtain a second prediction result. Among them, by adaptively switching the large model reasoning process, the large model is automatically used to reason about sensor data in short-term and abnormal conditions, meeting different reasoning and prediction requirements, and adaptively monitoring and predicting the production and preparation environment of stem cells in real time, ensuring the timeliness and reasoning depth of the large model reasoning, and effectively improving the timeliness of anomaly detection in the production and preparation environment of stem cells using large model reasoning.

[0026] Next, an exemplary explanation is given of the large-scale model reasoning method for stem cell production and preparation provided in this application.

[0027] like Figure 1 As shown, the large model reasoning method for stem cell production and preparation provided in this application includes the following steps: Step 11, obtaining sensor data of multiple sensors in the stem cell production and preparation Internet of Things at the current moment.

[0028] The above-mentioned sensor data are data of corresponding detection indicators collected by sensors in the Internet of Things for stem cell production and preparation. The corresponding detection indicators include temperature, humidity, pH value, carbon dioxide concentration, dissolved oxygen concentration and other indicators. The corresponding data are temperature data, humidity data, pH value data, carbon dioxide concentration data, dissolved oxygen concentration data and the like.

[0029] In an embodiment of the present application, the sensor data of multiple sensors at the current moment can be obtained by accessing the data uploaded in real time by sensors in the Internet of Things.

[0030] Exemplarily, the Internet of Things for stem cell production and preparation is an Internet of Things related to monitoring the production environment of stem cells, including monitoring equipment (sensors, video monitoring components, etc.), control systems (automated control systems), data transmission equipment (gateway equipment, wireless networks, etc.), management systems (systems that monitor and analyze relevant data of the stem cell production environment to achieve real-time monitoring, alarm, fault diagnosis and other functions, such as computers running large language models), and sensors are used to measure data of detection indicators for stem cell production and preparation, including temperature, humidity, pH value, etc.

[0031] Step 12: construct a first prompt template based on all sensor data at the current moment, and use the large model to infer the first prompt template to obtain a first prediction result.

[0032] The first prompt template is used to prompt the large model to perform abnormal detection on the detection index, and the first prediction result is used to describe whether the detection index corresponding to each sensor data at the current moment is abnormal. If the first prediction result is the label of the detection index corresponding to each sensor, when the label is 1, the corresponding detection index is abnormal, and when the label is 0, the corresponding detection index is normal. The above-mentioned large model is an abbreviation for large-scale pre-trained models (such as large language models (LLM)) in the field of artificial intelligence, and has strong logical reasoning capabilities.

[0033] In some embodiments of the present application, the first prompt template is constructed based on all sensor data at the current moment, and the first prompt template is inferred using a large model to obtain a first prediction result, including: The first step is to build the first prompt template based on all sensor data at the current moment.

[0034] Specifically, through the formula: Constructing the First Prompt Template .

[0035] in, represents the first target of the large model, represents the mode threshold, Represents all sensor data at the current moment, Represents the first domain knowledge.

[0036] It should be noted that the first goal is used to describe the task content that the large model is expected to complete (in this step, it is to classify and predict whether the sensor data is abnormal). The first domain knowledge is the professional knowledge or background information required for the large model to complete the first goal, which helps the large model provide more accurate and professional answers. The mode threshold is the reference range value of the sensor data in the current task scenario determined based on industry knowledge [ ],in is the minimum value of the range, is the maximum value of the range, and and is a matrix, which includes the minimum range value and the maximum range value corresponding to all sensor data; the + in the above formula is a splicing combination operation, which is used to splice all data into one piece of data.

[0037] Exemplarily, when the stem cell production and preparation Internet of Things is an Internet of Things related to monitoring the production environment of stem cells, the above-mentioned first goal is to classify the abnormalities of the detection indicators corresponding to each sensor in the stem cell production environment at the current moment, and the first domain knowledge is the professional knowledge such as the normal value range and variation range of the data of the detection indicators in the stem cell production and preparation scenario.

[0038] In the second step, the large model is used to infer the first prompt template to obtain the first prediction result.

[0039] Specifically, the large model converts the content in the first prompt template into multiple tokens (i.e., tokens, which are the smallest units for the large model to process and understand language, which can be a word, character, subword, or symbol), and maps all tokens into a first high-dimensional vector representation (which can be implemented using a mapping function), and then uses the attention mechanism to contextually encode the first high-dimensional vector representation, pass the intermediate representation layer by layer, and obtain the first encoding result. Finally, the first encoding result is decoded according to the attention weight (which can be implemented using the Transformer architecture) to obtain the first prediction result.

[0040] Exemplarily, the big model can be a large language model, etc. The first prompt template is input into the big model, and the big model processes the first prompt template to obtain a first prediction result. Specifically, the big model performs shallow reasoning on multiple sensor data streams based on the first target in the first prompt template and combined with the first domain knowledge, and outputs a prediction result.

[0041] It should be noted that in order to improve the accuracy of the first prediction result, the large model needs to be trained before actually obtaining the first prediction result at the current moment. The sensor data at the historical moment is used as training data, and the large model is used to obtain the first prediction result of the training data. Then, the online gradient loss is calculated based on the first prediction result of the training data. If the value of the online gradient loss is less than the preset loss value, the training ends. If the value of the online gradient loss is greater than or equal to the preset loss value, the model weight of the large model is adjusted, and the step of obtaining the first prediction result of the training data using the large model is returned. Specifically, through the formula: Calculate online loss value .

[0042] in, Indicates the first prediction result of the training data. The class probability distribution of sensor data, Represents the true prediction result of the training data The class probability distribution of sensor data, Indicates the number of sensor data in the training data.

[0043] By formula: Adjust the model weight of the large model to obtain the adjusted model weight .

[0044] in, represents the model weight of the large model, represents the learning rate, Represents the gradient.

[0045] This step is illustrated below with reference to a specific example.

[0046] The flow chart of the sentinel mode of the large model (i.e., the mode of executing this step to obtain the first prediction result) is as follows Figure 2 As shown, the sensor signal data Upload it to the prompt template, and load the prompt template content through the knowledge base to obtain the prompt template (that is, the first prompt template mentioned above). The prompt template is input into the big model, and the big model performs shallow reasoning to obtain the prediction result. , according to the prediction results, the prompt template is fed back and the threshold is modified, and according to the actual results and prediction results Update the large model's weights through online learning, and finally , , Save to knowledge base.

[0047] Step 13, when all sensor data at the current moment and the first prediction result meet the inference mode switching condition, a second prompt template is constructed based on all sensor data at the current moment and the output format of the large model, and the second prompt template is inferred using the large model to obtain a second prediction result.

[0048] The above-mentioned second prompt template is used to prompt the large model to predict abnormal conditions of the detection indicators. The second prediction result is used to describe the abnormal conditions of the detection indicators corresponding to each sensor at future moments. For example, for the sensor data of temperature in the stem cell production environment, the second prediction result is: within the next hour, the temperature sensor data in the stem cell production environment predicts that there is a 70% probability that the ambient temperature will fluctuate abnormally, and the temperature may rise sharply to above 39°C, which exceeds the ideal temperature range required for stem cell growth.

[0049] The above-mentioned reasoning mode switching condition is: the first prediction result describes that the detection indicator corresponding to at least one sensor data at the current moment is abnormal, or the value of at least one sensor data among all the sensor data at the current moment is not within the numerical range described by the mode threshold.

[0050] It should be noted that, since the mode threshold includes the numerical range corresponding to all sensor data, when determining whether the sensor data is within the numerical range described by the mode threshold, the sensor data is compared with the numerical range corresponding to the sensor data in the mode threshold.

[0051] In an embodiment of the present application, the steps of constructing a second prompt template based on all sensor data at the current moment and the output format of the large model, and using the large model to infer the second prompt template to obtain a second prediction result include: In the first step, the second prompt template is constructed based on all sensor data at the current moment and the output format of the large model.

[0052] Specifically, through the formula: Constructing the Second Prompt Template .

[0053] in, represents the second goal of the large model, represents the mode threshold, Represents all sensor data at the current moment, represents the second domain knowledge, Represents the output format for large models.

[0054] It should be noted that the second goal is used to describe the task content that the large model is expected to complete (in this step, it is to predict abnormal conditions of sensor data at future moments), the second domain knowledge is the professional knowledge or background information required by the large model to complete the second goal, and the output format is the format of the second prediction result. The + in the above formula is a splicing combination operation, which is used to splice all data into one piece of data.

[0055] Exemplarily, when the stem cell production and preparation Internet of Things is an Internet of Things related to production environment monitoring, the above-mentioned second goal can be to predict whether there will be abnormal fluctuations in the carbon dioxide concentration in the stem cell production environment in the next five minutes. The second domain knowledge can be the range of ideal environmental conditions required for stem cell growth and the mechanism of the impact of carbon dioxide concentration on stem cell growth. The output format can be the probability and confidence interval of abnormal fluctuations in the carbon dioxide concentration in the stem cell production environment in the next five minutes.

[0056] In the second step, the large model is used to infer the second prompt template to obtain the second prediction result.

[0057] Specifically, the large model converts the content in the second prompt template into multiple tokens, and maps all tokens into a second high-dimensional vector representation, then performs context encoding on the second high-dimensional vector representation through the attention mechanism, passes the intermediate representation layer by layer to obtain the second encoding result, and finally decodes the second encoding result according to the attention weight to obtain the second prediction result.

[0058] Exemplarily, the second prompt template is input into the big model, and the big model processes the second prompt template to obtain a second prediction result. Specifically, the big model infers abnormal conditions that may occur in the future based on the second target in the second prompt template and the required second domain knowledge by performing zero-sample reasoning on the sensor data.

[0059] It should be noted that after this step, the model threshold needs to be updated. Specifically, the formula is: Update the mode threshold to obtain the updated mode threshold .

[0060] in, Indicates the lower limit of the numerical range described by the mode threshold. Indicates the upper limit of the numerical range described by the mode threshold. Represents all sensor data at the current moment.

[0061] The updated mode threshold is used for calculation at the next moment after the current moment.

[0062] It should be noted that updating the mode threshold according to the sensor data at the current moment can improve the adaptability of the mode threshold to the current task.

[0063] It is worth mentioning that by actively using the big model to infer the abnormal conditions of the sensor data when the mode switching conditions are met, it is inferred that abnormal conditions may occur in the environment of stem cell production and preparation in the short term in the future, and alarms are issued based on the abnormal conditions, so that the environment of stem cell production and preparation can be intervened in advance to reduce losses.

[0064] This step is described below with reference to a specific example.

[0065] The early warning mode of the large model (i.e., the mode of executing this step and obtaining the second prediction result) is as follows Figure 3 As shown, the sensor data is uploaded to the prompt template, and the knowledge base is used to reorganize the prompt template content to obtain a second prompt template. The second prompt template is input into the large model for zero-sample reasoning to obtain the reasoning result (i.e., the second prediction result mentioned above), and the reasoning result is used to provide feedback and threshold correction to the prompt template.

[0066] In some embodiments of the present application, when the termination moment of the reasoning cycle is reached, the sensor data of all sensors at each moment in the reasoning cycle are used to generate examples for each moment, a third prompt template is constructed based on all the examples, and the third prompt template is inferred using a large model to obtain a long-term reasoning prediction result.

[0067] The above long-term reasoning prediction results are used to describe the data changes of the detection indicators within the reasoning cycle. For example, for the pH value, the long-term reasoning prediction results are: within the reasoning cycle, the pH value of the culture medium shows a trend of first decreasing and then stabilizing. At the beginning of the cycle, due to the gradual increase in cell metabolic activity, more metabolites such as carbon dioxide are produced, causing the pH value of the culture medium to gradually decrease from the initial 7.4 to about 7.2. As the production cycle progresses, the cells gradually adapt to the growth environment, and the metabolic balance tends to stabilize. At the same time, the environmental control system's regulatory effect on the CO2 concentration becomes apparent, which slows down the rate of pH decrease and eventually fluctuates within the range of 7.2-7.3, maintaining a relatively stable state. This pH value change trend meets the physiological needs of stem cell growth and is conducive to the continuous proliferation and differentiation of cells.

[0068] For example, the reasoning cycle is 30 minutes, and the end time of the previous reasoning cycle is 8:00. If the current time is 8:20, the end time of the reasoning cycle has not been reached. If the current time is 8:30, the end time of the reasoning cycle has been reached. For the reasoning cycle from 8:00 to 8:30, the multiple times may be: 8:00, 8:10, 8:20, 8:30.

[0069] In an embodiment of the present application, the steps of using sensor data of all sensors at each moment in the reasoning cycle to generate examples at each moment, constructing a third prompt template according to all examples, and using a large model to reason on the third prompt template to obtain a long-term reasoning prediction result include: In the first step, the sensor data of all sensors at each moment in the inference cycle are utilized to generate examples at each moment.

[0070] Specifically, through the formula: Generate Examples of moments .

[0071] in, Indicates the problem target, Indicates The triplet corresponding to each moment is , Indicates a moment, Indicates All sensor data at a moment in time, Indicates The actual result of the moment, , represents the reasoning chain, Indicates answer.

[0072] It should be noted that the problem goal is used to describe the task that needs to be completed in the example, the reasoning chain is used to describe the steps, basis, and logical relationship involved in the reasoning process of the example to get the answer based on the problem goal and the triple, and the answer is used to describe the conclusion drawn in the example based on the problem goal and the reasoning chain. For example, when the Internet of Things for stem cell production and preparation is an Internet of Things related to production environment monitoring, the above problem goal can be to analyze the changes in ambient temperature during the reasoning cycle and determine whether there are abnormal fluctuations. The reasoning chain can be to first extract the ambient temperature data at each moment in the reasoning cycle, calculate the average, maximum, minimum, standard deviation and other statistical indicators of the ambient temperature, identify abnormal temperature fluctuations based on the statistical indicators of the ambient temperature, and finally analyze the causes and potential risks of abnormal temperature fluctuations in combination with the operating status and historical data of the environmental control system. The answer can be that during the reasoning cycle, the average temperature of the stem cell production environment is 37.2°C, the maximum is 38.5°C, and the minimum is 36.5°C. There is an abnormal situation of large temperature fluctuations, which may be related to the lag in adjustment of the environmental control system or changes in the external ambient temperature. It is recommended to check the sensors and adjustment equipment of the environmental control system and optimize the temperature control parameters to ensure the stability of the ambient temperature.

[0073] In the second step, a third prompt template is constructed based on all examples, and the third prompt template is inferred using the large model to obtain long-term reasoning prediction results.

[0074] Specifically, through the formula: Constructing the Third Prompt Template .

[0075] in, Indicates the third goal, represents the mode threshold, represents the set of triples corresponding to all moments, Represents the third domain knowledge, Represents the set of examples corresponding to all moments.

[0076] It should be noted that the + in the above formula is a concatenation operation, which is used to concatenate all data into one piece of data. The above third goal is used to describe the task content that the large model is expected to complete (in this step, it is to analyze the data changes of the detection indicators during the reasoning cycle), and the third field knowledge is the professional knowledge or background information required for the large model to complete the third goal.

[0077] Exemplarily, when the stem cell production and preparation Internet of Things is an Internet of Things related to production environment monitoring, the above-mentioned third goal is to analyze the changing trend of environmental conditions in the stem cell production environment during the reasoning cycle to evaluate the stability and reliability of the environmental control system and provide a basis for optimizing the environmental control strategy. The third field of knowledge includes the influence of changes in environmental conditions on the growth quality of stem cells, the adjustment mechanism of the environmental control system and the fault diagnosis method.

[0078] Then the third prompt template is input into the large model for reasoning to obtain the long-term reasoning prediction result.

[0079] For example, the large model is based on the prompt template Medium Target The large model first inputs Extract the corresponding time node ,entity , and relationship , and construct the evidence quadruple ( , , , ), by using the example Learning to generate explanation hints. Evidence quadruple ( , , , ) and explanation hints as the reasoning process of the knowledge chain. Then the big model is constructed by analyzing the time nodes in the input data X. and real results , Evidence quadruple ( , , , ) and other inductive summaries to conduct deep reasoning on the knowledge chain and output long-term reasoning prediction results.

[0080] It should be noted that the third goal is used to describe the task content that the large model is expected to complete (in this step, it is to detect changes in sensor data at all times), and the third domain knowledge is used to describe the expertise or background information required for the large model to complete the third goal.

[0081] This step is illustrated below with reference to a specific example.

[0082] The detection mode of the large model (i.e., the mode of executing this step to obtain long-term inference prediction results) is as follows Figure 4As shown, examples 1, 2, ..., k are input into the prompt template, together with the target, domain knowledge, pattern threshold and input sensor data to form a third prompt template. During the reasoning process, the third prompt template is reasoned based on the evidence quadruple and the explanation prompt to obtain the reasoning answer (i.e., the long-term reasoning prediction result mentioned above).

[0083] It is worth mentioning that by adaptively switching the large model reasoning process, the large model can automatically reason about sensor data in the short term and abnormal conditions, meet different reasoning and prediction needs, and adaptively monitor and predict anomalies in the stem cell production and preparation environment in real time, ensuring the timeliness and reasoning depth of the large model reasoning, and effectively improving the timeliness of anomaly detection in the stem cell production and preparation environment using large model reasoning.

[0084] In addition, the method of the present application relies on a prompt template framework that includes a pattern threshold, and controls the adaptive switching of the large model through multiple modes and reasoning depths to achieve monitoring and reasoning prediction of data from multiple periods. Through incremental fine-tuning of online learning, online learning is performed on real-time data streams, and the weights and pattern thresholds of the large model are continuously adjusted to adapt them to the personalized needs of specific tasks. The knowledge chain enhanced by time series knowledge combines time series knowledge and associations to build a knowledge chain to reason and analyze long-term trends, thereby improving the accuracy of long-term predictions. Utilizing the powerful reasoning ability and rich world knowledge of the large model, through adaptive switching modes and online learning, real-time monitoring and personalized analysis of data can be achieved, and the accuracy and robustness of reasoning predictions can be improved. Through multiple modes and reasoning depths, the reasoning and prediction needs for time periods of different lengths can be met, such as short-term anomaly detection, long-term trend analysis, and other needs.

[0085] The following is an exemplary description of the large-scale model reasoning device for stem cell production and preparation provided in this application.

[0086] like Figure 5 As shown, the embodiment of the present application provides a large model reasoning device for stem cell production and preparation, and the large model reasoning device 500 for stem cell production and preparation includes: An acquisition module 501 is used to acquire sensor data of multiple sensors in the stem cell production and preparation Internet of Things at the current moment; The first inference module 502 is used to construct a first prompt template based on all sensor data at the current moment, and use the large model to infer the first prompt template to obtain a first prediction result; the first prediction result is used to describe whether the detection index corresponding to each sensor at the current moment is abnormal, and the first prompt template is used to prompt the large model to perform abnormal detection on the detection index; The second reasoning module 503 is used to construct a second prompt template based on all sensor data at the current moment and the output format of the large model when all sensor data at the current moment and the first prediction result meet the reasoning mode switching condition, and use the large model to infer the second prompt template to obtain a second prediction result; the second prediction result is used to describe the abnormal condition of the detection indicator corresponding to each sensor at a future moment, and the second prompt template is used to prompt the large model to predict the abnormal condition of the detection indicator.

[0087] It should be noted that the information interaction, execution process, etc. between the above-mentioned devices / units are based on the same concept as the method embodiment of the present application. Their specific functions and technical effects can be found in the method embodiment part and will not be repeated here.

[0088] The technicians in the relevant field can clearly understand that for the convenience and simplicity of description, only the division of the above-mentioned functional units and modules is used as an example for illustration. In practical applications, the above-mentioned function allocation can be completed by different functional units and modules as needed, that is, the internal structure of the device can be divided into different functional units or modules to complete all or part of the functions described above. The functional units and modules in the embodiment can be integrated in a processing unit, or each unit can exist physically separately, or two or more units can be integrated in one unit. The above-mentioned integrated unit can be implemented in the form of hardware or in the form of software functional units. In addition, the specific names of the functional units and modules are only for the convenience of distinguishing each other, and are not used to limit the scope of protection of this application. The specific working process of the units and modules in the above-mentioned system can refer to the corresponding process in the aforementioned method embodiment, which will not be repeated here.

[0089] like Figure 6 As shown, an embodiment of the present application provides a terminal device. The terminal device D10 of this embodiment includes: at least one processor D100 ( Figure 6 Only one processor is shown in the figure), a memory D101, and a computer program D102 stored in the memory D101 and executable on the at least one processor D100, wherein the processor D100 implements the steps of any of the above-mentioned method embodiments when executing the computer program D102.

[0090] Specifically, when the processor D100 executes the computer program D102, it obtains the sensor data of multiple sensors at the current moment in the stem cell production and preparation Internet of Things, then builds a first prompt template based on all the sensor data at the current moment, and uses the large model to infer the first prompt template to obtain a first prediction result. When all the sensor data at the current moment and the first prediction result meet the inference mode switching condition, a second prompt template is built based on all the sensor data at the current moment and the output format of the large model, and the large model is used to infer the second prompt template to obtain a second prediction result. Among them, by adaptively switching the large model inference process, the large model automatically infers the sensor data in the short term and abnormal conditions, meets different inference and prediction requirements, and realizes adaptive real-time monitoring and abnormal prediction of the stem cell production and preparation environment, ensuring the timeliness and inference depth of the large model inference, and effectively improving the timeliness of abnormal detection of the stem cell production and preparation environment using large model inference.

[0091] The processor D100 may be a central processing unit (CPU), or other general-purpose processors, digital signal processors (DSP), application-specific integrated circuits (ASIC), field-programmable gate arrays (FPGA), or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc. A general-purpose processor may be a microprocessor or any conventional processor, etc.

[0092] In some embodiments, the memory D101 may be an internal storage unit of the terminal device D10, such as a hard disk or memory of the terminal device D10. In other embodiments, the memory D101 may also be an external storage device of the terminal device D10, such as a plug-in hard disk, a smart memory card (SMC, SmartMedia Card), a secure digital (SD, Secure Digital) card, a flash card (Flash Card), etc. equipped on the terminal device D10. Further, the memory D101 may also include both an internal storage unit of the terminal device D10 and an external storage device. The memory D101 is used to store an operating system, an application program, a boot loader (BootLoader), data, and other programs, such as the program code of the computer program, etc. The memory D101 may also be used to temporarily store data that has been output or is to be output.

[0093] An embodiment of the present application further provides a computer-readable storage medium, wherein the computer-readable storage medium stores a computer program, and when the computer program is executed by a processor, the steps in the above-mentioned method embodiments can be implemented.

[0094] An embodiment of the present application provides a computer program product. When the computer program product runs on a terminal device, the terminal device can implement the steps in the above-mentioned method embodiments when executing the computer program product.

[0095] If the integrated unit is implemented in the form of a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the present application implements all or part of the processes in the above-mentioned embodiment method, which can be completed by instructing the relevant hardware through a computer program, and the computer program can be stored in a computer-readable storage medium. When the computer program is executed by the processor, the steps of the above-mentioned method embodiments can be implemented. Among them, the computer program includes computer program code, and the computer program code can be in source code form, object code form, executable file or some intermediate form. The computer-readable medium may at least include: any entity or device, recording medium, computer memory, read-only memory (ROM, Read-Only Memory), random access memory (RAM, Random Access Memory), electric carrier signal, telecommunication signal and software distribution medium that can carry the computer program code to the large model reasoning method device / terminal device for stem cell production and preparation. For example, a USB flash drive, a mobile hard disk, a magnetic disk or an optical disk. In some jurisdictions, according to legislation and patent practice, computer-readable media cannot be electric carrier signals and telecommunication signals.

[0096] In the above embodiments, the description of each embodiment has its own emphasis. For parts that are not described or recorded in detail in a certain embodiment, reference can be made to the relevant descriptions of other embodiments.

[0097] Those of ordinary skill in the art will appreciate that the units and algorithm steps of each example described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware or a combination of computer software and electronic hardware. Whether these functions are performed in hardware or software depends on the specific application and design constraints of the technical solution. Professional and technical personnel can use different methods to implement the described functions for each specific application, but such implementation should not be considered to be beyond the scope of this application.

[0098] The above is a preferred embodiment of the present application. It should be pointed out that for ordinary technicians in this technical field, several improvements and modifications can be made without departing from the principles described in the present application. These improvements and modifications should also be regarded as the scope of protection of the present application.

Claims

1. A large-scale model reasoning method for stem cell production and preparation, characterized in that: include: Obtain sensor data of multiple sensors in the stem cell production and preparation Internet of Things at the current moment; A first prompt template is constructed based on all sensor data at the current moment, and the first prompt template is inferred using the large model to obtain a first prediction result; the first prediction result is used to describe whether the detection index corresponding to each sensor at the current moment is abnormal, the first prompt template is used to prompt the large model to perform abnormal detection on the detection index, the content in the first prompt template is converted into a first high-dimensional vector representation using the large model, and the first high-dimensional vector representation is context-encoded and then decoded to obtain a first prediction result; When all the sensor data at the current moment and the first prediction result meet the inference mode switching condition, a second prompt template is constructed based on all the sensor data at the current moment and the output format of the large model, and the large model is used to infer the second prompt template to obtain a second prediction result; the second prediction result is used to describe the abnormal condition of the detection indicator corresponding to each sensor at a future moment, and the second prompt template is used to prompt the large model to predict the abnormal condition of the detection indicator, and the large model is used to convert the content in the second prompt template into a second high-dimensional vector representation, and the second high-dimensional vector representation is context encoded and then decoded to obtain a second prediction result.

2. The large model reasoning method according to claim 1, characterized in that: The step of constructing a first prompt template based on all sensor data at the current moment includes: By formula: Constructing the First Prompt Template ; in, represents the first target of the large model, represents the mode threshold, Represents all sensor data at the current moment, represents the first domain knowledge, and + represents the concatenation and combination operation.

3. The large model reasoning method according to claim 1, characterized in that: The reasoning mode switching condition is: The first prediction result describes that at the current moment, there is at least one sensor data corresponding to an abnormal detection indicator, or at least one sensor data among all the sensor data at the current moment has a value that is not within the numerical range described by the mode threshold.

4. The large model reasoning method according to claim 1, characterized in that: The second prompt template is constructed based on all sensor data at the current moment and the output format of the large model, including: By formula: Constructing the Second Prompt Template ; in, represents the second goal of the large model, represents the mode threshold, Represents all sensor data at the current moment, represents the second domain knowledge, Indicates the output format of the large model, and + indicates the concatenation operation.

5. The large model reasoning method according to claim 1, characterized in that: The large model reasoning method also includes: When the termination moment of the reasoning cycle arrives, the sensor data of all sensors at each moment in the reasoning cycle are used to generate examples for each moment, a third prompt template is constructed based on all the examples, and the third prompt template is inferred using a large model to obtain a long-term reasoning prediction result; the long-term reasoning prediction result is used to describe the data changes of the detection indicators within the reasoning cycle.

6. The large model reasoning method according to claim 5, characterized in that: The using the sensor data of all sensors at each moment in the reasoning cycle to generate an example at each moment includes: By formula: Generate Examples of moments ; in, Indicates the problem target, Indicates The triplet corresponding to each moment is , Indicates a moment, Indicates the All sensor data at a moment in time, Indicates the The actual result of the moment, , represents the reasoning chain, Indicates answer, + indicates concatenation operation; The third prompt template is constructed according to all the examples, including: By formula: Constructing the Third Prompt Template ; in, Indicates the third goal, represents the mode threshold, represents the set of triples corresponding to all moments, Represents the third domain knowledge, Represents the set of examples corresponding to all moments.

7. The large model reasoning method according to claim 3, characterized in that: After the step of using the large model to infer the second prompt template to obtain a second prediction result, the large model inference method further includes: By formula: Update the mode threshold to obtain the updated mode threshold ; in, Indicates the lower limit of the numerical range described by the mode threshold. represents the upper limit of the numerical range described by the mode threshold, Represents all sensor data at the current moment.

8. A large-scale model inference device based on stem cell production, characterized in that: include: An acquisition module, used to acquire sensor data of multiple sensors in the stem cell production and preparation Internet of Things at the current moment; A first reasoning module is used to construct a first prompt template based on all sensor data at the current moment, and use the large model to infer the first prompt template to obtain a first prediction result; the first prediction result is used to describe whether the detection index corresponding to each sensor at the current moment is abnormal, the first prompt template is used to prompt the large model to perform abnormal detection on the detection index, and the content in the first prompt template is converted into a first high-dimensional vector representation by using the large model, and the first high-dimensional vector representation is context-encoded and then decoded to obtain a first prediction result; The second reasoning module is used to construct a second prompt template based on all sensor data at the current moment and the output format of the large model when all sensor data at the current moment and the first prediction result meet the reasoning mode switching condition, and use the large model to infer the second prompt template to obtain a second prediction result; the second prediction result is used to describe the abnormal condition of the detection indicator corresponding to each sensor at a future moment, the second prompt template is used to prompt the large model to predict the abnormal condition of the detection indicator, and the large model is used to convert the content in the second prompt template into a second high-dimensional vector representation, and the second high-dimensional vector representation is context-encoded and then decoded to obtain a second prediction result.

9. A terminal device, comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that: When the processor executes the computer program, the large model reasoning method as described in any one of claims 1 to 7 is implemented.

10. A computer-readable storage medium storing a computer program, characterized in that: When the computer program is executed by a processor, the large model reasoning method as described in any one of claims 1 to 7 is implemented.

Images