A medicine metal foreign matter online intelligent detection method and detection machine

Abstract

The present application belongs to the technical field of medicine metal foreign matter detection, and in particular relates to a kind of medicine metal foreign matter online intelligent detection method and detection machine, the method comprises: through array probe coil, the electromagnetic induction signal of each column medicine is collected, the suspected rate of metal foreign matter is calculated, according to the suspected rate, the medicine row is divided into suspected medicine row and ordinary row, single medicine in suspected foreign matter conveying structure is handled by overturning, multi-angle electromagnetic induction signal is collected and the confirmation rate of metal foreign matter is calculated, and according to this, unqualified medicine is screened and marked, in the rejection station, the accurate rejection of unqualified medicine is realized by baffle, the total number of medicine and the number of unqualified medicine are counted simultaneously, and the final qualified rate of medicine is calculated in combination with the confirmation rate of metal foreign matter.The present application improves the accuracy and efficiency of online detection by the detection mode of multi-column parallel conveying and preliminary screening followed by accurate inspection, and guarantees the quality of medicine production.

Description

Technical Field

[0001] This invention belongs to the technical field of metal foreign object detection in pharmaceuticals, and specifically relates to an online intelligent detection method and detection machine for metal foreign objects in pharmaceuticals. Background Technology

[0002] Metal foreign objects in pharmaceuticals can endanger patients' medication safety. These foreign objects may originate from: wear and tear on production equipment, contamination of raw materials, mixing in during the packaging process, and human error. To prevent pharmaceuticals containing metal foreign objects from entering the market, it is necessary to conduct online detection of metal foreign objects in pharmaceuticals.

[0003] Existing online detection methods for metallic foreign objects in pharmaceuticals typically use electromagnetic induction to determine the presence of metallic foreign objects. These methods mostly involve batch testing, resulting in low detection efficiency and high costs. Furthermore, during the testing process, the pharmaceuticals are transported in a fixed position, and metallic foreign objects inside the pharmaceuticals may create blind spots due to their different positions and orientations, making it difficult to achieve comprehensive detection and leading to a high rate of missed detections.

[0004] Furthermore, existing technologies have relatively simple methods for processing electromagnetic signals, typically relying solely on the amplitude and phase of the electromagnetic signal for judgment. This makes it difficult to distinguish between interference from the drug itself and abnormal signals from metallic foreign objects, thus failing to accurately detect and identify drugs containing metallic foreign objects, resulting in low accuracy in foreign object identification. Summary of the Invention

[0005] In view of this, in order to solve the above problems, an online intelligent detection method and detection machine for metallic foreign objects in pharmaceuticals are proposed.

[0006] The objective of this invention can be achieved through the following technical solution: This invention provides an online intelligent detection method for metallic foreign objects in pharmaceuticals. The method includes: laying the pharmaceuticals to be tested in multiple columns and conveying them to the detection station; collecting electromagnetic induction signals of each column of pharmaceuticals through an array probe coil; and calculating the suspected rate of metallic foreign objects in each column of pharmaceuticals.

[0007] Based on the suspected metal foreign object rate, each drug row is divided into a suspected drug row and the rest of the drug rows. The two are then diverted to their respective single-column conveying structures by a diversion mechanism, which are denoted as the suspected foreign object conveying structure and the normal conveying structure, respectively.

[0008] Individual drugs within the suspected foreign object transport structure are flipped over, and electromagnetic induction signals from various angles are collected one by one by the detection probe coil to calculate the metal foreign object confirmation rate of each drug.

[0009] Based on the metal foreign object confirmation rate, unqualified drugs are screened and marked. When an unqualified drug is transported to the rejection station of the corresponding suspected foreign object transport structure, the control system controls the baffle of the rejection station to open. The baffle is immediately closed after the unqualified drug falls.

[0010] Record and tally the total number of drugs and the number of substandard drugs, and calculate the final pass rate of drugs by combining the metal foreign object confirmation rate of individual drugs.

[0011] Compared with the prior art, the beneficial effects of the present invention are as follows: (1) The present invention uses a multi-column parallel conveying and a detection method of initial screening followed by fine inspection. First, the array probe coil is used to perform initial screening on each drug column, and then only the suspected columns are diverted for precise detection of a single drug. While ensuring the continuous operation of the production line, the abnormality is located to a single drug, thereby improving the detection efficiency and accurately locating a single unqualified drug.

[0012] (2) By performing multi-angle flipping detection on suspected drugs and performing positive and negative assignment cancellation processing on the electromagnetic feature signals corresponding to complementary symmetrical angles, this invention not only eliminates the detection blind zone caused by the fixed position of the drug and the random position of the metal foreign object, but also cancels the inherent reference interference caused by the structure of the drug itself and the geometric symmetry of the coil, thereby reducing the false alarm rate and improving the all-round detection and stability.

[0013] (3) This invention performs weighted fusion and normalization quantization processing based on multi-dimensional electromagnetic characteristics of signal amplitude, phase, equivalent impedance and equivalent inductance. It uses two-level quantization judgment parameters of suspected metal foreign object rate and confirmed metal foreign object rate. Compared with the traditional single signal judgment method, the accuracy of foreign object identification and anti-interference ability are significantly improved. It can accurately distinguish between small metal foreign objects and drug interference itself, and improve detection accuracy. Attached Figure Description

[0014] To more clearly illustrate the technical solutions of the embodiments of the present invention, the accompanying drawings used in the description of the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0015] Figure 1 This is a schematic diagram of the overall implementation process of the present invention.

[0016] Figure 2 This is a schematic diagram illustrating the calculation process for the suspected metal foreign object rate in this invention.

[0017] Figure 3 This is a schematic diagram illustrating the calculation process for the metal foreign object identification rate of this invention. Detailed Implementation

[0018] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0019] This invention is based on the scenario of online intelligent detection of metal foreign objects in the pharmaceutical manufacturing process. Specifically, the detection machine used in this invention can be used with any tablet press or capsule filling machine, or it can operate independently as a standalone detection device, adapting to the online detection needs of various pharmaceutical products such as tablets and capsules. When the detection machine is used with a tablet press or capsule filling machine, the pharmaceutical product directly enters the detection process of this invention after the preceding tableting or capsule filling process, without the need for additional transfer. When used independently, the pharmaceutical product is conveyed to the detection equipment through a feeding mechanism.

[0020] After the medicines enter the testing machine, they are laid out in multiple evenly spaced rows. Limiting components are installed on the surface of the conveying structure to laterally separate the rows of medicines, ensuring they remain neat and do not overlap during transport. Above the conveying path of the multiple rows of medicines, array probe coils are arranged in a matrix corresponding to the number of rows. Each row of medicines corresponds to an independent set of detection probe coils. The array probe coils span the entire conveying channel of the multiple rows of medicines, maintaining a fixed detection distance from the conveying structure below.

[0021] For a detailed analysis of specific implementation methods related to this scenario, please refer to [link / reference]. Figure 1 As shown, this invention provides an online intelligent detection method for metallic foreign objects in pharmaceuticals. The method includes: S1, laying the pharmaceuticals to be tested in multiple rows and conveying them to the detection station; collecting electromagnetic induction signals from each row of pharmaceuticals using an array probe coil; and calculating the suspected metallic foreign object rate for each row of pharmaceuticals. Preferably, the multiple rows are three or more, and the spacing between the pharmaceuticals laid out in each row is greater than or equal to 2 cm.

[0022] Specifically, please refer to Figure 2 As shown, the calculation process for the suspected metal foreign object rate includes: extracting the signal amplitude and signal phase from the electromagnetic induction signal; and calculating the equivalent impedance and equivalent inductance of the probe coil based on the excitation frequency, excitation voltage, and coil internal resistance of the array probe coil, combined with the signal amplitude and signal phase.

[0023] Specifically, the calculation process of the equivalent impedance includes: calculating the phase difference between the electromagnetic induction signal and the excitation signal, and combining the coil internal resistance to calculate the equivalent impedance.

[0024] The formula for calculating equivalent impedance is: In the formula, For equivalent impedance, The coil internal resistance is the excitation frequency of the array probe coil. The phase difference between the electromagnetic induction signal and the excitation signal. The value is the cosine of the phase difference between the electromagnetic induction signal and the excitation signal, representing the ratio of the resistive component to the equivalent impedance.

[0025] Specifically, the calculation process of the equivalent inductance includes: calculating the equivalent inductance based on the coil internal resistance, excitation frequency, and phase difference between the electromagnetic induction signal and the excitation signal.

[0026] The formula for calculating equivalent inductance is: In the formula, For equivalent inductance, The coil internal resistance is the excitation frequency of the array probe coil. The phase difference between the electromagnetic induction signal and the excitation signal. For the excitation frequency, The tangent of the phase difference between the electromagnetic induction signal and the excitation signal represents the ratio of the inductive reactance to the coil internal resistance in the detection circuit. Inductive reactance is the impedance of an inductor to alternating current. ω is the angular frequency of the alternating excitation signal.

[0027] The reference signal amplitude, phase, excitation frequency, excitation voltage, and coil resistance are extracted from the reference electromagnetic induction signal. Similarly, the reference equivalent impedance and inductance are calculated using the same methods as for calculating equivalent impedance and inductance changes. The reference electromagnetic induction signal is the standard reference electromagnetic induction signal corresponding to a drug without metallic foreign objects.

[0028] The amplitude difference, phase difference, equivalent impedance difference, and equivalent inductance difference between the electromagnetic induction signal and the reference electromagnetic induction signal are calculated respectively. The absolute values ​​of each difference are weighted and summed to obtain a comprehensive characteristic value. Then, the value is subjected to minimum-maximum linear normalization to obtain the suspected metal foreign object rate.

[0029] In the electromagnetic induction detection of metallic foreign objects in pharmaceuticals, the signal amplitude difference is the most direct result of the electromagnetic disturbance caused by the metallic foreign object, and therefore it is given the highest weight. The equivalent impedance difference comprehensively reflects the combined influence of the probe coil excitation parameters, the electromagnetic characteristics of the pharmaceutical product, and the electromagnetic response of the coil by the metallic foreign object. The stronger the anti-interference capability, the higher the detection stability, and it is an auxiliary feature for foreign object identification, so it is given the second highest weight. The phase difference can distinguish between the material of the metallic foreign object and the electromagnetic interference of the pharmaceutical product itself, improving the identification accuracy, and is given a medium weight. The equivalent inductance difference mainly reflects the change in permeability and is used to supplement the verification of ferromagnetic foreign objects, playing a corrective and fault-tolerant role, so it is given the lowest weight. As a preferred embodiment of the present invention, the weights of the amplitude difference, phase difference, equivalent impedance difference, and equivalent inductance difference are 0.4, 0.2, 0.3, and 0.1, respectively.

[0030] The formula for calculating the comprehensive eigenvalue is: In the formula, For comprehensive eigenvalues, The difference in amplitude. It is the absolute value of the amplitude difference. The phase difference value. It is the absolute value of the phase difference. This is the equivalent impedance difference. It is the absolute value of the equivalent impedance difference. The difference in equivalent inductance. It is the absolute value of the difference in equivalent inductance.

[0031] It should be added that the maximum and minimum values ​​of the comprehensive characteristic values ​​are selected from the comprehensive characteristic values ​​of all the drugs listed, and are used as the maximum and minimum values ​​required for the normalization, respectively. The minimum-maximum linear normalization process is an existing technology and will not be described in detail in this invention.

[0032] S2. Based on the suspected metal foreign object rate, each drug row is divided into a suspected drug row and the rest of the drug rows. The two are then introduced into the corresponding single-column conveying structure through a diversion mechanism, which are respectively denoted as the suspected foreign object conveying structure and the normal conveying structure.

[0033] Specifically, the process of dividing the suspected drug column from the other drug columns includes: classifying the drug columns with a metal foreign object suspicion rate greater than or equal to a preset safety threshold as suspected drug columns.

[0034] It should be noted that the above-mentioned preset safety threshold can be obtained based on industry experience values ​​for electromagnetic induction detection of foreign metal objects in pharmaceuticals. Preferably, the preset safety threshold is 0.3.

[0035] The remaining drug rows are denoted as the remaining drug rows.

[0036] S3. For individual medications within the suspected foreign object transport structure, flip them over. Electromagnetic induction signals from various angles are collected one by one using the detection probe coil to calculate the metal foreign object confirmation rate for each medication. Then, the flipping structures and detection probe coils of the remaining medication rows are closed, and the medications in the remaining rows are transported directly.

[0037] Specifically, the electromagnetic induction signal acquisition process at each angle is as follows: A flipping structure is set in the middle of the suspected foreign object conveying structure. The flipping angle of the flipping structure is set to 0 degrees to 360 degrees. The flipping structure flips at preset flipping angle intervals, causing the medicine to rotate. Signal acquisition is triggered at the preset flipping angle position. Preferably, the preset flipping angle interval is 90 degrees, and signal acquisition is triggered at four angle positions: 0 degrees, 90 degrees, 180 degrees, and 270 degrees.

[0038] A plurality of pairs of detection probe coils are symmetrically arranged on the flipping structure, with the central axis of the detection probe coils perpendicular to the central axis of the drug flipping. Preferably, the plurality of pairs is one or more.

[0039] The detection probe coil is electrically connected to the control system. When the medicine is flipped to each preset flipping angle, the detection probe coil collects the electromagnetic induction signal of each medicine at each preset flipping angle in real time.

[0040] Specifically, please refer to Figure 3 As shown, the calculation process of the metal foreign object confirmation rate includes: extracting the signal amplitude and signal phase of the electromagnetic induction signal at each preset flip angle; and calculating the equivalent impedance and equivalent inductance corresponding to each preset flip angle based on the excitation frequency, excitation voltage and coil internal resistance of the probe coil, in the same way as the calculation method of the equivalent impedance and equivalent inductance of each drug.

[0041] Based on the shape of the medicine, the preset flipping angles that are relatively opposite and geometrically symmetrical during the flipping process are denoted as complementary symmetric flipping angles.

[0042] Assign a positive sign to the equivalent impedance corresponding to any angle in the complementary symmetric flip angle, and a negative sign to the equivalent impedance corresponding to its symmetrical angle, and calculate the sum of the equivalent impedances for all angles.

[0043] Similarly, the total equivalent inductance can be calculated using the same method as the total equivalent impedance.

[0044] The absolute values ​​of the sum of equivalent impedances and the sum of equivalent inductances are weighted and summed to obtain the comprehensive foreign matter index. This index is then subjected to minimum-maximum linear normalization to obtain the metal foreign matter confirmation rate of the corresponding drug.

[0045] Equivalent impedance is more sensitive to the eddy current loss of metallic foreign objects, has a stronger signal, and is less affected by the drug matrix and environmental interference. It is the main basis for judging the presence of metallic foreign objects, and therefore is given a higher weight. In a preferred embodiment of the present invention, the weights of the absolute value of the sum of equivalent impedances and the absolute value of the sum of equivalent inductances are set to 0.7 and 0.3, respectively.

[0046] It should be added that the selection method for the maximum and minimum values ​​required for normalizing the above-mentioned comprehensive foreign matter index is the same as the selection method for the maximum and minimum values ​​required for normalizing the comprehensive feature value.

[0047] The locations of metallic foreign objects inside the drug are randomly distributed. Using only a single angle to collect electromagnetic induction signals can easily lead to missed detections because the foreign object is in the detection blind zone. At the same time, there is a geometric symmetry between the detection probe coil and the drug, and the drug's own structure will generate inherent reference signal interference.

[0048] This invention acquires circumferential electromagnetic induction signals by multi-angle flipping and assigns positive and negative signs to the equivalent impedances corresponding to the complementary symmetrical flipping angles, which can cancel the inherent reference signals caused by the geometric symmetry of the coil and the structure of the drug itself. This preserves the real abnormal eddy current signals generated by the metallic foreign objects, avoiding detection blind spots caused by the random position of foreign objects and misjudgments caused by interference from geometric structures and reference signals. This reduces the false alarm rate and improves detection sensitivity and stability. At the same time, flipping and precise detection are only activated for suspected drug rows, while other drug rows are directly transported. This ensures that the overall detection and transport efficiency is not reduced while maintaining detection accuracy.

[0049] S4. Based on the metal foreign object confirmation rate, unqualified drugs are screened and marked. When an unqualified drug is transported to the rejection station of the corresponding suspected foreign object transport structure, the control system controls the baffle of the rejection station to open. The baffle is closed immediately after the unqualified drug falls.

[0050] Specifically, the screening process for non-compliant drugs includes marking drugs with a metal foreign object detection rate greater than or equal to a preset detection rate threshold as non-compliant drugs.

[0051] The process of obtaining the aforementioned preset confirmation rate threshold is as follows: collect multiple batches of qualified drugs without metal foreign objects, calculate the mean and standard deviation of their metal foreign object confirmation rate, and use the sum of the mean and three times the standard deviation as the preset confirmation rate threshold. Preferably, the multiple batches are 30 batches.

[0052] The location information of substandard drugs is recorded and synchronized to the control system.

[0053] The remaining drugs are marked as qualified drugs and transported to the next workstation.

[0054] Pharmaceutical production lines typically involve multiple parallel and high-speed continuous conveying. If each drug is flipped and precisely inspected at multiple angles, the conveying speed will be reduced, while the inspection time and system computing load will be increased, making it difficult to adapt to the needs of online high-speed production. If the entire column is inspected uniformly, small foreign objects in individual drugs within the column will be masked by the overall signal, resulting in missed detection of tiny foreign objects and making it impossible to locate specific non-conforming drugs.

[0055] This invention first performs rapid initial screening by drug rows, enabling the rapid release of large batches of qualified drugs. Then, it performs detailed inspection on individual drugs in drug rows suspected of containing foreign objects, which can pinpoint the anomaly to a specific individual drug and avoid missed detections and misjudgments caused by averaging the signal of the entire row. This approach balances the detection efficiency of the production line with the accuracy of identifying metal foreign objects, thus avoiding situations where the entire row is qualified but a single drug is unqualified and cannot be identified.

[0056] S5. Record and count the total number of drugs and the number of unqualified drugs. Combine the metal foreign object confirmation rate of each drug to calculate the final pass rate of the drugs.

[0057] Specifically, the calculation process of the final pass rate includes: calculating the difference between the total number of drugs and the number of unqualified drugs to obtain the number of qualified drugs.

[0058] Calculate the ratio of the number of qualified drugs to the total number of drugs, and record it as the basic qualification rate.

[0059] Calculate the average metal foreign body detection rate of qualified drugs, and use the difference between 1 and the average metal foreign body detection rate as the quality correction factor.

[0060] The final pass rate is obtained by multiplying the quality correction factor by the baseline pass rate.

[0061] The present invention also provides an online intelligent detection machine for metal foreign objects in pharmaceuticals, including a detection machine body and a processing unit and control system built into the detection machine body. The detection machine body is provided with a conveying structure, an array probe coil is provided on the detection machine body and above the conveying structure, and a position sensor is provided on the conveying structure and directly below the array probe coil. The array probe coil and the position sensor are respectively communicatively connected to the processing unit; wherein, as a preferred embodiment of the present invention, the conveying structure is a conveyor belt.

[0062] A diversion mechanism is set on the main body of the detection machine and located to the right of the array probe coil. Multiple suspected foreign object conveying structures and multiple ordinary conveying structures are linearly and evenly arranged on the front and back sides to the right of the diversion mechanism.

[0063] A flipping mechanism is provided in the middle of the suspected foreign object conveying structure. A detection probe coil is provided above the suspected foreign object conveying structure and above the flipping mechanism. The detection probe coil is communicatively connected to the processing unit. A baffle is provided on the suspected foreign object conveying structure and to the right of the flipping mechanism. The baffle is controlled to open and close by the control system.

[0064] When the testing machine is working, the drugs to be tested, laid out in rows, are first transported to the testing station via a conveyor structure. When the drugs reach below the array probe coil, a position sensor triggers signal acquisition. At this time, the array probe coil collects the electromagnetic induction signal corresponding to each row of drugs and transmits it to the processing unit. The processing unit calculates the suspected metal foreign object rate for each row of drugs. Then, the diversion mechanism divides each row of drugs into suspected drug rows and other drug rows based on the suspected metal foreign object rate, and imports them into the corresponding suspected foreign object conveying structure and the normal conveying structure. The flipping structure in the suspected foreign object conveying structure then processes the individual drugs... The medicines are flipped over, and electromagnetic induction signals from various angles during the flipping process are collected one by one by a detection probe coil. These signals are then sent to the processing unit, which calculates the metal foreign object confirmation rate for each medicine, filters and marks unqualified medicines. When an unqualified medicine is transported to the rejection station of the corresponding suspected foreign object conveying structure, the processing unit sends a signal to the control system to open the baffle. At this time, the control system opens the baffle to allow the unqualified medicine to fall off, and immediately closes the baffle after the unqualified medicine falls off. The remaining medicines are then transported to the next station. Ordinary conveying structures directly transport medicines to the next station.

[0065] The above content is merely an example and illustration of the concept of the present invention. Those skilled in the art can make various modifications or additions to the specific embodiments described, or use similar methods to replace them, as long as they do not deviate from the concept of the invention or exceed the scope defined by the present invention, and all such modifications and additions should fall within the protection scope of the present invention.

Claims

1. An online intelligent detection method for metallic foreign objects in pharmaceuticals, characterized in that, The method includes: The drugs to be tested are laid out in multiple columns and transported to the testing station. The electromagnetic induction signals of each column of drugs are collected by the array probe coil, and the suspected rate of metal foreign objects in each column of drugs is calculated. Based on the suspected metal foreign object rate, each drug row is divided into a suspected drug row and the rest of the drug rows. The two are then introduced into the corresponding single-column conveying structure through a diversion mechanism, which are respectively denoted as the suspected foreign object conveying structure and the normal conveying structure. The individual medicines in the suspected foreign object transport structure are flipped over, and the electromagnetic induction signals at each angle are collected one by one by the detection probe coil to calculate the metal foreign object confirmation rate of the individual medicines. Based on the metal foreign object confirmation rate, unqualified drugs are screened and marked. When an unqualified drug is transported to the rejection station of the corresponding suspected foreign object transport structure, the control system controls the baffle of the rejection station to open. The baffle is immediately closed after the unqualified drug falls. Record and count the total number of drugs and the number of unqualified drugs, and calculate the final pass rate of drugs by combining the metal foreign object confirmation rate of individual drugs. The calculation process for the suspected metallic foreign object rate includes: The signal amplitude and signal phase are extracted from the electromagnetic induction signal. Based on the excitation frequency, excitation voltage and coil internal resistance of the array probe coil, the equivalent impedance and equivalent inductance of the probe coil are calculated by combining the signal amplitude and signal phase. The reference signal amplitude, signal phase, excitation frequency and excitation voltage, and coil internal resistance are extracted from the reference electromagnetic induction signal. The reference equivalent impedance and reference equivalent inductance are calculated in the same way as the calculation method of equivalent impedance and inductance change. The amplitude difference, phase difference, equivalent impedance difference, and equivalent inductance difference between the electromagnetic induction signal and the reference electromagnetic induction signal are calculated respectively. The absolute values ​​of each difference are weighted and summed to obtain a comprehensive characteristic value. Then, the value is subjected to minimum-maximum linear normalization to obtain the suspected metal foreign object rate.

2. The online intelligent detection method for metallic foreign objects in pharmaceuticals as described in claim 1, characterized in that: The calculation process for the equivalent impedance includes: Calculate the phase difference between the electromagnetic induction signal and the excitation signal, and combine it with the coil internal resistance to calculate the equivalent impedance.

3. The online intelligent detection method for metallic foreign objects in pharmaceuticals as described in claim 2, characterized in that: The calculation process for the equivalent inductance includes: The equivalent inductance is calculated based on the coil internal resistance, excitation frequency, and phase difference between the electromagnetic induction signal and the excitation signal.

4. The online intelligent detection method for metallic foreign objects in pharmaceuticals as described in claim 1, characterized in that: The process of separating the suspected drug category from the other drug categories includes: Drugs with a suspected metal foreign object rate greater than or equal to a preset safety threshold are classified as suspected drugs. The remaining drug rows are denoted as the remaining drug rows.

5. The online intelligent detection method for metallic foreign objects in pharmaceuticals as described in claim 1, characterized in that: The process of acquiring the electromagnetic induction signals at each angle is as follows: A flipping structure is set in the middle of the suspected foreign object transport structure. The flipping angle of the flipping structure is set from 0 degrees to 360 degrees. It flips at preset flipping angle intervals. The flipping structure drives the medicine to rotate and triggers signal acquisition at the preset flipping angle position. Several pairs of detection probe coils are symmetrically arranged on the flipping structure, and the central axis of the detection probe coils is perpendicular to the central axis of the drug flipping structure. The detection probe coil is electrically connected to the control system. When the medicine is flipped to each preset flipping angle, the detection probe coil collects the electromagnetic induction signal of each medicine at each preset flipping angle in real time.

6. The online intelligent detection method for metallic foreign objects in pharmaceuticals as described in claim 5, characterized in that: The calculation process for the metal foreign object confirmation rate includes: The signal amplitude and signal phase of the electromagnetic induction signal at each preset flip angle are extracted respectively. Based on the excitation frequency, excitation voltage and coil internal resistance of the probe coil, the equivalent impedance and equivalent inductance corresponding to each preset flip angle are calculated in the same way as the calculation method of the equivalent impedance and equivalent inductance of each drug. Based on the shape of the medicine, the preset flipping angles that are relatively opposite and geometrically symmetrical during the flipping process are denoted as complementary symmetric flipping angles. Assign a positive sign to the equivalent impedance corresponding to any angle in the complementary symmetric flip angle, and a negative sign to the equivalent impedance corresponding to its symmetrical angle. Calculate the sum of the equivalent impedances for all angles. Similarly, the total equivalent inductance can be calculated using the same method as the total equivalent impedance. The absolute values ​​of the sum of equivalent impedances and the sum of equivalent inductances are weighted and summed to obtain the comprehensive foreign matter index. This index is then subjected to minimum-maximum linear normalization to obtain the metal foreign matter confirmation rate of the corresponding drug.

7. The online intelligent detection method for metallic foreign objects in pharmaceuticals as described in claim 1, characterized in that: The screening process for the substandard drugs includes: Drugs with a metal foreign object detection rate greater than or equal to a preset detection rate threshold are marked as unqualified drugs. The location information of substandard drugs is recorded and synchronized to the control system. The remaining drugs are marked as qualified drugs and transported to the next workstation.

8. The online intelligent detection method for metallic foreign objects in pharmaceuticals as described in claim 1, characterized in that: The calculation process for the final pass rate includes: Calculate the difference between the total number of drugs and the number of substandard drugs to obtain the number of qualified drugs; Calculate the ratio of the number of qualified drugs to the total number of drugs, and record it as the basic pass rate; Calculate the average metal foreign body detection rate of qualified drugs, and use the difference between 1 and the average metal foreign body detection rate as the quality correction factor; The final pass rate is obtained by multiplying the quality correction factor by the baseline pass rate.

9. An online intelligent detection machine for metal foreign objects in pharmaceuticals, used to perform the online intelligent detection method for metal foreign objects in pharmaceuticals as described in any one of claims 1-8, characterized in that: The testing machine includes a main body and a processing unit and control system built into the main body. The main body is provided with a conveying structure. An array probe coil is provided on the main body and above the conveying structure. A position sensor is provided on the conveying structure and directly below the array probe coil. The array probe coil and the position sensor are respectively communicatively connected to the processing unit. The conveying structure is a conveyor belt. A diversion mechanism is set on the main body of the detection machine and located to the right of the array probe coil. Multiple suspected foreign object conveying structures and multiple ordinary conveying structures are linearly and evenly arranged on the front and rear sides to the right of the diversion mechanism. A flipping mechanism is provided in the middle of the suspected foreign object conveying structure. A detection probe coil is provided above the suspected foreign object conveying structure and above the flipping mechanism. The detection probe coil is communicatively connected to the processing unit. A baffle is provided on the suspected foreign object conveying structure and to the right of the flipping mechanism. The baffle is controlled to open and close by the control system.

Images