Intelligent medical equipment power supply shell fault early warning system and method

Abstract

The invention belongs to the technical field of intelligent medical equipment, and particularly relates to an intelligent medical equipment power supply shell fault early warning system and method, and the system specifically comprises the following modules: a power supply module, a shell module, a distributed sensor network module, an information fusion module, and a remote monitoring module. The power supply module comprises a voltage regulation module, a current control module, a power distribution module and a fault protection module, the distributed sensor network module is installed outside the shell module and used for monitoring state information of temperature, pressure, vibration and deformation in real time, each sensor carries out binary decision through a local detector, and the fault protection module carries out fault protection. And the information fusion module fuses the received binary decision, namely local decision results, and diagnoses the local decision through an improved particle filter algorithm to obtain a fault detection result to trigger the fault protection module to carry out early warning. The influence of single-dimension misjudgment on a final result is reduced, and the accuracy of fault judgment is further improved.

Description

TECHNICAL FIELD

[0001] The application belongs to the technical field of intelligent medical equipment, and particularly relates to an intelligent medical equipment power shell fault early warning system and method. BACKGROUND

[0002] In the medical field, the stable operation of intelligent medical equipment is of great importance, which is directly related to the life safety of patients and the quality of medical services. As an important protective structure of the power supply part of intelligent medical equipment, the state of the power shell has an important influence on the safe operation of the equipment. The power shell may have temperature abnormalities, excessive vibration, deformation, surface damage and other faults due to long-term use, environmental factors, external force collision and other reasons. If these faults cannot be discovered and handled in time, the internal components of the power supply may be damaged, which may cause equipment downtime, electrical leakage and other safety hazards, and in severe cases, it may even endanger the safety of patients and medical staff.

[0003] At present, the fault detection of the power shell of intelligent medical equipment mainly relies on the monitoring data of a single sensor, and these methods have limitations in dealing with complex systems, such as large communication volume, poor real-time performance, and inability to effectively process nonlinear and non-Gaussian systems. With the continuous development of intelligent medical technology, there is an urgent need for a system and method that can real-time and accurately warn of power shell faults. SUMMARY

[0004] The application provides an intelligent medical equipment power shell fault early warning system and method, which solves the technical problems of single and limited fault monitoring of the power shell of existing intelligent medical equipment.

[0005] In order to achieve the above purpose, the technical scheme adopted by the application is as follows: an intelligent medical equipment power shell fault early warning system, specifically comprising the following modules: a power module, a shell module, a distributed sensor network module, an information fusion module and a remote monitoring module, the power module includes a voltage regulation module, a current control module, a power distribution module and a fault protection module, the shell module is used to protect the power module from external interference, the distributed sensor network module is installed outside the shell module and is used to monitor the state information of temperature, pressure, vibration and deformation in real time, each sensor makes a binary decision through a local detector, the information fusion module fuses the received binary decision, i.e. local decision result, and diagnoses the local decision through an improved particle filtering algorithm to obtain a fault detection result and transmit the power module, triggering the fault protection module to perform early warning, and the remote control module is used to receive fault analysis results and early warning signals and store and display them.

[0006] Furthermore, the voltage regulation module ensures that the output voltage is stable within the range required by the device, the current control module limits the current output to prevent overload, the power distribution module rationally distributes power to each power system, and the fault protection module automatically cuts off the power supply when an abnormality is detected to protect the device safety.

[0007] Furthermore, the housing module is used to protect the power module and internal components from external interference. The distributed sensor network module is installed outside the housing module and includes temperature sensors, pressure sensors, vibration sensors, and deformation sensors. The states of the sensors can be described by formulas.

[0008] Furthermore, the sensors of the distributed sensor network module perform binary decision-making through local detectors, compare the detection information with a threshold, and output a local decision. The local detector is a type of sensor that can determine whether a local fault has occurred in the outer casing module.

[0009] Furthermore, the normal state of the power supply casing fault early warning system is as follows: The fault status is The dynamic system satisfies the Markov property, that is: The state of the system at a certain moment depends only on the state at the previous moment, where, for Time's up The state transition probability density function at time t, assuming it has The system is detected by a state sensor, and the state equation and observation equation of the dynamic system are: , ;in These represent two states: normal and faulty. For process noise, This refers to the measurement noise of the sensor.

[0010] Furthermore, the expression for the local detector to determine whether a fault has occurred is as follows: ,in, At any moment No. The measurement values ​​from each sensor are the thresholds for determining local faults. These are local decision values ​​that need to be transmitted to the diagnostic center. It is binary information.

[0011] Furthermore, the information fusion processor fuses all the local binary decision information after receiving it, and gives the final decision result according to the following formula:

[0012] ,in, It is a statistical function responsible for counting time points. All fault information, The system fault decision threshold at time k; Indicates time The system fault decision result is 0, indicating that the system is working normally, and 1, indicating that the system is in a fault state. If the system is determined to be in a fault state, the system responds by triggering the fault protection module. If the system is determined to be working normally, the local posterior probability density functions are updated. .

[0013] Furthermore, the information fusion processor diagnoses the local decision using an improved particle filter algorithm as follows:

[0014] Step 1: Perform the following sub-steps for each of the m status monitoring sensors:

[0015] 1) According to the formula Estimate the probability of observation;

[0016] 2) Based on the joint particle representation at time k-1 and ;

[0017] 3) According to calculate ;

[0018] Step 2: Calculation ;

[0019] Step 3: According to calculate ;

[0020] Step 4: If It is 0, according to the formula Where N is the total number of particles. Let be the weight of particle i. ,in For importance functions, can be used Instead of updating the posterior probability and weights, output a warning message.

[0021] A method comprising a power supply casing fault early warning system for an intelligent medical device includes the following steps:

[0022] S1: The housing module is used to protect the power module from external interference. A distributed sensor network module is installed outside the housing module to monitor the status information of temperature, pressure, vibration and deformation in real time. The status information includes temperature, pressure, deformation and deformation.

[0023] S2: The distributed sensor network module performs binary decision-making through local detectors;

[0024] S3: The information fusion module fuses the received binary decision, i.e., local decision results, and simultaneously diagnoses the local decisions through an improved particle filter algorithm, obtains fault detection results, and transmits them to the power module;

[0025] S4: Trigger the fault protection module in the power module to issue an early warning. The remote control module is used to receive the fault analysis results and early warning signals, and to store and display them.

[0026] Furthermore, the power supply module includes a voltage regulation module, a current control module, a power distribution module, and a fault protection module.

[0027] Compared with the prior art, the present invention has the following beneficial effects:

[0028] (1) The fault detection results of the present invention can be directly transmitted to the power module to form a closed-loop link of "diagnosis-transmission-protection". After optimization, the computational efficiency of the improved particle filter algorithm can quickly complete the diagnosis process, reduce decision delay, and ensure that the power module can trigger the fault protection module immediately after receiving the results.

[0029] (2) This invention integrates multi-state monitoring sensors on the power supply casing, compares the data information collected by the sensors with the threshold through the 0,1 binary decision method, and finally gives the decision result through the information fusion processor to determine whether it is in a fault state. If it is confirmed to be in a fault state, a fault warning is given. It can integrate the binary judgment results of various dimensions, perform cross-validation and comprehensive analysis, reduce the impact of single-dimensional misjudgment on the final result, and further improve the accuracy of fault judgment.

[0030] (3) This invention can run quickly without the need for a high-performance computing chip. The multi-state monitoring sensors are mostly mature conventional sensors that are easy to acquire and install, which reduces the overall cost of the system and makes it easier to promote and apply to the power supply casings of different types of smart medical devices. Attached Figure Description

[0031] Figure 1 This is a schematic diagram of the power supply casing fault early warning system of the present invention. Detailed Implementation

[0032] The present invention will now be described in detail with reference to specific embodiments. These embodiments will help those skilled in the art to further understand the present invention, but do not limit the invention in any way. It should be noted that those skilled in the art can make several changes and improvements without departing from the concept of the present invention. These all fall within the protection scope of the present invention.

[0033] like Figure 1As shown, an intelligent medical device power supply casing fault early warning system specifically includes the following modules: a power supply module, a casing module, a distributed sensor network module, an information fusion module, and a remote monitoring module. The power supply module includes a voltage regulation module, a current control module, a power distribution module, and a fault protection module. The casing module is used to protect the power supply module from external interference. The distributed sensor network module is installed outside the casing module and is used to monitor the status information of temperature, pressure, vibration, and deformation in real time. Each sensor performs binary decision-making through a local detector. The information fusion module fuses the received binary decisions, i.e., local decision results, and simultaneously diagnoses the local decisions through an improved particle filter algorithm to obtain fault detection results and transmit them to the power supply module, triggering the fault protection module to issue an early warning. The remote control module is used to receive fault analysis results and early warning signals, and to store and display them.

[0034] The power module is one of the core components of intelligent medical devices, responsible for providing stable power support. Its main functions include: a voltage regulation module to ensure the output voltage remains stable within the required range; a current control module to limit current output and prevent overload; a power distribution module to rationally allocate power to various power systems; and a fault protection module to automatically cut off power when an anomaly is detected, protecting the device.

[0035] The enclosure protects the power module and other internal components from external interference, such as dust, liquids, and electromagnetic interference. The enclosure design must meet the following requirements:

[0036] Protection rating: Depending on the operating environment of the medical device, the enclosure should meet a certain protection rating, such as IP65 or higher; Heat dissipation performance: Ensure that the heat generated by the power module during operation can be effectively dissipated to prevent overheating; Mechanical strength: The enclosure should have sufficient mechanical strength to withstand a certain amount of external impact.

[0037] A distributed sensor network (DSN) consists of multiple condition monitoring sensors installed at key locations on the power supply housing module to monitor condition information such as temperature, pressure, vibration, and deformation in real time. Each sensor performs binary decision-making through a local detector, comparing the detected information with a threshold and outputting local decision binary information.

[0038] The state of each sensor can be described by the following formula: ,in Let h be the measurement value of the l-th sensor at time k, and h be the observation function. It is the system status. It is observation noise, assuming .

[0039] The mathematical expression for the local detector is as follows: ,in, ) is the measurement value of the l-th sensor at time k. It is the threshold for judging local faults. These are local decision values ​​that need to be passed to the fault protection module.

[0040] The normal state of the power supply casing fault warning system is as follows: The fault status is The dynamic system satisfies the Markov property, that is: The state of the system at a certain moment depends only on the state at the previous moment, where, for Time's up The state transition probability density function at time t, assuming it has The system is detected by a state sensor, and the state equation and observation equation of the dynamic system are: , ;in These represent two states: normal and faulty. For process noise, This refers to the measurement noise of the sensor.

[0041] Each monitoring sensor detects the state of various parts of the power module, and the acquired data is processed by local detectors for binary decision-making. The information fusion processor then fuses the received binary decision results to derive the fault detection result and issue an early warning. Since the data transmitted from each local detector to the information fusion processor is only binary information, the amount of communication is reduced, as is the computational load of information fusion, thereby improving the efficiency of fault diagnosis and enabling real-time fault alarms.

[0042] The expression for determining whether a fault has occurred by a local detector is:

[0043] ,in, At any moment No. The measurement values ​​from each sensor are the thresholds for determining local faults. These are local decision values ​​that need to be transmitted to the diagnostic center. It is binary information. The information fusion processor fuses all the local binary decision information received and gives the final decision result according to the following formula:

[0044] ,in, It is a statistical function responsible for counting time points. All fault information, The system fault decision threshold at time k; Indicates time The system fault decision result is 0, indicating that the system is working normally, and 1, indicating that the system is in a fault state. If the system is determined to be in a fault state, the system responds by triggering the fault protection module. If the system is determined to be working normally, the local posterior probability density functions are updated. .

[0045] Assuming that the measurement data of each sensor are independent and identically distributed, the observation probability density function of each sensor under normal and fault conditions can be expressed by the following formula.

[0046] (1)

[0047] Here, e=0 and 1 represent normal and fault scenarios, respectively. For local decision-making and threshold optimization, it is assumed that the decision rules for each sensor are the same. This is the threshold for local decision-making, and the integration region corresponding to local decision-making is divided into... So the range of internal normal and fault conditions can be...

[0048] By separating pairs and In the region Points are earned through internal requests.

[0049] For the fusion decision error expression:

[0050] ,in and They are respectively and Prior probabilities in two scenarios and These are two cases of statistical probability, where the function Q(g) is the complementary distribution function of the standard Gaussian function. For random variables The mean, Let be the standard deviation of the random variable.

[0051] To reduce errors, the log-likelihood method is used to improve the local decision and fusion decision as follows:

[0052] Local decision expression:

[0053] (2)

[0054] in This is the threshold for local decision-making.

[0055] Fusion decision expression:

[0056] (3)

[0057] in The threshold for fusion decision.

[0058] Based on the above analysis, an improved particle filter algorithm is presented. The process by which the information fusion processor uses the improved particle filter algorithm to diagnose the local decisions is as follows:

[0059] Step 1: Perform the following sub-steps for each of the m status monitoring sensors:

[0060] 1) Estimate the observation probability according to equation (1);

[0061] 2) Based on the joint particle representation at time k-1 and ;

[0062] 3) Calculate according to formula (2) ;

[0063] Step 2: Calculation ;

[0064] Step 3: Calculate according to equation (3) ;

[0065] Step 4: If It is 0, according to the formula Where N is the total number of particles. Let be the weight of particle i. ,in For importance functions, can be used Instead of updating the posterior probability and weights, output a warning message.

[0066] The warning results are transmitted to the power module, triggering the fault protection module to issue a warning. The remote control module receives the fault analysis results and warning signals, and stores and displays them.

[0067] A method for a power supply casing fault early warning system for intelligent medical devices includes the following steps:

[0068] S1: The housing module is used to protect the power module from external interference. A distributed sensor network module is installed outside the housing module to monitor the status information of temperature, pressure, vibration and deformation in real time. The status information includes temperature, pressure, deformation and deformation.

[0069] S2: The distributed sensor network module performs binary decision-making through local detectors;

[0070] S3: The information fusion module fuses the received binary decision, i.e., local decision results, and simultaneously diagnoses the local decisions through an improved particle filter algorithm, obtains fault detection results, and transmits them to the power module;

[0071] S4: Trigger the fault protection module in the power module to issue an early warning. The remote control module is used to receive the fault analysis results and early warning signals, and to store and display them.

[0072] The data transmitted from each local detector to the information fusion processor in this invention is only binary information, which reduces the amount of communication and the amount of computation for information fusion, improves the efficiency of fault diagnosis, and provides the possibility for real-time fault alarm.

[0073] Specific embodiments of the present invention have been described above. It should be understood that the present invention is not limited to the specific embodiments described above, and those skilled in the art can make various changes or modifications within the scope of the claims, which do not affect the essence of the present invention. Unless otherwise specified, the embodiments and features described in this application can be arbitrarily combined with each other.

Claims

1. A fault early warning system for the power supply casing of an intelligent medical device, characterized in that, Specifically, it includes the following modules: a power supply module, a housing module, a distributed sensor network module, an information fusion module, and a remote monitoring module. The power supply module includes a voltage regulation module, a current control module, a power distribution module, and a fault protection module. The housing module is used to protect the power supply module from external interference. The distributed sensor network module is installed outside the housing module and is used to monitor the status information of temperature, pressure, vibration, and deformation in real time. Each sensor performs binary decision-making through a local detector. The information fusion module fuses the received binary decision-making results, i.e., the local decision-making results, and diagnoses the local decisions through an improved particle filter algorithm to obtain fault detection results and transmit them to the power supply module, triggering the fault protection module to issue an early warning. The remote control module is used to receive fault analysis results and early warning signals, and to store and display them.

2. The intelligent medical device power supply casing fault early warning system according to claim 1, characterized in that, The voltage regulation module ensures that the output voltage is stable within the range required by the equipment, the current control module limits the current output to prevent overload, the power distribution module rationally distributes power to each power system, and the fault protection module automatically cuts off the power supply to protect the equipment safety when an abnormality is detected.

3. The intelligent medical device power supply casing fault early warning system according to claim 1, characterized in that, The housing module is used to protect the power module and internal components from external interference. The distributed sensor network module is installed outside the housing module and includes temperature sensors, pressure sensors, vibration sensors, and deformation sensors. The state of the sensors can be described by formulas.

4. The intelligent medical device power supply casing fault early warning system according to claim 1, characterized in that, The sensors in the distributed sensor network module perform binary decision-making through local detectors, compare the detection information with a threshold, and output a local decision. The local detector is a type of sensor that can determine whether a local fault has occurred in the outer casing module.

5. The intelligent medical device power supply casing fault early warning system according to claim 4, characterized in that, The power supply casing fault early warning system is in normal condition as follows: The fault status is The dynamic system satisfies the Markov property, that is: The state of the system at a certain moment depends only on the state at the previous moment, where, for Time's up The state transition probability density function at time t, assuming it has The system is detected by a state sensor, and the state equation and observation equation of the dynamic system are: , ;in These represent two states: normal and faulty. For process noise, This refers to the measurement noise of the sensor.

6. The intelligent medical device power supply casing fault early warning system and method according to claim 5, characterized in that, The expression used by the local detector to determine whether a fault has occurred is: ,in, At any moment No. The measurement values ​​from each sensor are the thresholds for determining local faults. These are local decision values ​​that need to be transmitted to the diagnostic center. It is binary information.

7. The intelligent medical device power supply casing fault early warning system and method according to claim 6, characterized in that, The information fusion processor fuses all the local binary decision information received and gives the final decision result according to the following formula: ,in, It is a statistical function responsible for counting time points. All fault information, The system fault decision threshold at time k; Indicates time The system fault decision result is 0, indicating that the system is working normally, and 1, indicating that the system is in a fault state. If the system is determined to be in a fault state, the system responds by triggering the fault protection module. If the system is determined to be working normally, the local posterior probability density functions are updated. .

8. The intelligent medical device power supply casing fault early warning system and method according to claim 7, characterized in that, The information fusion processor uses an improved particle filter algorithm to diagnose the local decision as follows: Step 1: Perform the following sub-steps for each of the m status monitoring sensors: 1) According to the formula Estimate the probability of observation; 2) Based on the joint particle representation at time k-1 and ; 3) According to calculate ; Step 2: Calculation ; Step 3: According to calculate ; Step 4: If It is 0, according to the formula Where N is the total number of particles. Let be the weight of particle i. ,in For importance functions, can be used Instead of updating the posterior probability and weights, output a warning message.

9. An optimized method comprising the intelligent medical device power supply casing fault early warning system according to any one of claims 1-8, characterized in that, Specifically, it includes the following steps: S1: The housing module is used to protect the power module from external interference. A distributed sensor network module is installed outside the housing module to monitor the status information of temperature, pressure, vibration and deformation in real time. The status information includes temperature, pressure, deformation and deformation. S2: The distributed sensor network module performs binary decision-making through local detectors; S3: The information fusion module fuses the received binary decision, i.e., local decision results, and simultaneously diagnoses the local decisions through an improved particle filter algorithm, obtains fault detection results, and transmits them to the power module; S4: Trigger the fault protection module in the power module to issue an early warning. The remote control module is used to receive the fault analysis results and early warning signals, and to store and display them.

10. An optimization method for a power supply casing fault early warning system for intelligent medical devices according to claim 9, characterized in that, The power supply module includes a voltage regulation module, a current control module, a power distribution module, and a fault protection module.

Images