Systems and methods for providing security in power systems

{
  "publication_number": "US10686810B1",
  "country": "US",
  "kind": "B1",
  "title": "Systems and methods for providing security in power systems",
  "abstract": "Systems and methods for security in power systems are provided. A security-aware distributed control framework for resilient operation of power systems can detect and mitigate different types of attacks that might target power systems. The framework can discover a change in the features of transmitted data from neighbor agents, discard an infected agent, and achieve an updated consensus protocol agreement while satisfying a control system objective.",
  "claims": [
    "1. A framework for providing security in a power system, the framework comprising: a hardware processor; and a machine-readable medium in operable communication with the hardware processor and having instructions stored thereon that, when executed by the hardware processor, perform the following steps: receive incoming information states from agents of the power system; analyze the incoming information states using a first level of a multi-resolution morphological gradient algorithm (MMGA); calculate a first resolution multi-resolution morphological gradient (MMG) based on the incoming information states; compare an absolute value of the first resolution MMG to a first threshold and, if the absolute value of the first resolution MMG exceeds the first threshold, trigger an attack alarm that the power system has an infected agent; if the attack alarm is triggered, calculate a second resolution MMG based on the incoming information states; compare an absolute value of the second resolution MMG to a second threshold to identify which of the agents of the power system is the infected agent; set all weighting factors of the infected agent to zero to exclude the infected agent from power system communication; and send an updated state to the agents of the power system with the infected agent excluded from the power system communication.",
    "2. The framework according to claim 1, the analyzing of the incoming information states using the first level of the MMGA comprising: calculating dilation and erosion processes of the incoming information states by finding a maximum number of delayed samples and a minimum number of delayed samples of the incoming information states.",
    "3. The framework according to claim 2, a quantity of the delayed samples being dependent upon a rate of information exchange between the agents of the power system.",
    "4. The framework according to claim 2, the calculating of the dilation and erosion processes comprising using the following equations: (f ⊕ g) ⁢ (k) = max ⁢ { f ⁡ (k + s) + g ⁡ (s) ❘ (k + s) ∈ D f, s ∈ D g } (f ⊖ g) ⁢ (k) = min ⁢ { f ⁡ (k + s) + g ⁡ (s) ❘ (k + s) ∈ D f, s ∈ D g }, where f represents a signal of the incoming information states, g is a structuring element that lies in domains D f, D g, s is a sampling constant of the signal, and k is a sampling constant of the structuring element.",
    "5. The framework according to claim 4, the calculating of the first resolution MMG comprising subtracting a dilated signal from an eroded signal.",
    "6. The framework according to claim 5, the subtracting of the dilated signal from the eroded signal comprising using the following equation: ∇ i w =(f⊕g) w −(f⊖g) w, where w represents a resolution level.",
    "7. The framework according to claim 2, the calculating of the first resolution MMG comprising subtracting a dilated signal from an eroded signal.",
    "8. The framework according to claim 7, the subtracting of the dilated signal from the eroded signal comprising using the following equation: ∇ i w =(f⊕g) w −(f⊖g) w, where w represents a resolution level.",
    "9. The framework according to claim 1, the calculating of the first resolution MMG comprising subtracting a dilated signal from an eroded signal.",
    "10. The framework according to claim 9, the subtracting of the dilated signal from the eroded signal comprising using the following equation: ∇ i w =(f⊕g) w −(f⊖g) w, where w represents a resolution level.",
    "11. The framework according to claim 1, the second threshold being greater than the first threshold.",
    "12. The framework according to claim 1, the first threshold being based on a topology of a communication graph between the agents of the power system, and the second threshold being based on the topology of the communication graph between the agents of the power system.",
    "13. The framework according to claim 1, the power system comprising at least one of a nanogrid, a microgrid, and a power system area.",
    "14. The framework according to claim 1, the agents of the power system comprising at least 20 agents.",
    "15. A method for providing security in a power system, the method comprising: receiving, by a hardware processor, incoming information states from agents of the power system; analyzing, by the hardware processor, the incoming information states using a first level of a multi-resolution morphological gradient algorithm (MMGA); calculating, by the hardware processor, a first resolution multi-resolution morphological gradient (MMG) based on the incoming information states; comparing, by the hardware processor, an absolute value of the first resolution MMG to a first threshold and, if the absolute value of the first resolution MMG exceeds the first threshold, trigger an attack alarm that the power system has an infected agent; if the attack alarm is triggered, calculating, by the hardware processor, a second resolution MMG based on the incoming information states; comparing, by the hardware processor, an absolute value of the second resolution MMG to a second threshold to identify which of the agents of the power system is the infected agent; setting, by the hardware processor, all weighting factors of the infected agent to zero to exclude the infected agent from power system communication; and sending, by the hardware processor, an updated state to the agents of the power system with the infected agent excluded from power system communication.",
    "16. The method according to claim 15, the analyzing of the incoming information states using the first level of the MMGA comprising: calculating dilation and erosion processes of the incoming information states by finding a maximum number of delayed samples and a minimum number of delayed samples of the incoming information states, the calculating of the dilation and erosion processes comprising using the following equations: (f ⊕ g) ⁢ (k) = max ⁢ { f ⁡ (k + s) + g ⁡ (s) ❘ (k + s) ∈ D f, s ∈ D g } (6) (f ⊖ g) ⁢ (k) = min ⁢ { f ⁡ (k + s) + g ⁡ (s) ❘ (k + s) ∈ D f, s ∈ D g }, (7) where f represents a signal of the incoming information states, g is a structuring element that lies in domains D f, D g, s is a sampling constant of the signal, and k is a sampling constant of the structuring element.",
    "17. The method according to claim 15, the calculating of the first resolution MMG comprising subtracting a dilated signal from an eroded signal, and the subtracting of the dilated signal from the eroded signal comprising using the following equation: ∇ i w =(f⊕g) w −(f⊖g) w, where w represents a resolution level.",
    "18. The method according to claim 15, the second threshold being greater than the first threshold, the first threshold being based on a topology of a communication graph between the agents of the power system, and the second threshold being based on the topology of the communication graph between the agents of the power system.",
    "19. The method according to claim 15, the power system comprising at least one of a nanogrid, a microgrid, and a power system area.",
    "20. A framework for providing security in a power system, the framework comprising: a hardware processor; and a machine-readable medium in operable communication with the hardware processor and having instructions stored thereon that, when executed by the hardware processor, perform the following steps: receive incoming information states from agents of the power system; analyze the incoming information states using a first level of a multi-resolution morphological gradient algorithm (MMGA); calculate a first resolution multi-resolution morphological gradient (MMG) based on the incoming information states; compare an absolute value of the first resolution MMG to a first threshold and, if the absolute value of the first resolution MMG exceeds the first threshold, trigger an attack alarm that the power system has an infected agent; if the attack alarm is triggered, calculate a second resolution MMG based on the incoming information states; compare an absolute value of the second resolution MMG to a second threshold to identify which of the agents of the power system is the infected agent; set all weighting factors of the infected agent to zero to exclude the infected agent from power system communication; and sending an updated state to the agents of the power system with the infected agent excluded from power system communication, the analyzing of the incoming information states using the first level of the MMGA comprising calculating dilation and erosion processes of the incoming information states by finding a maximum number of delayed samples and a minimum number of delayed samples of the incoming information states, a quantity of the delayed samples being dependent upon a rate of information exchange between the agents of the power system, the calculating of the dilation and erosion processes comprising using the following equations: (f ⊕ g) ⁢ (k) = max ⁢ { f ⁡ (k + s) + g ⁡ (s) ❘ (k + s) ∈ D f, s ∈ D g } (f ⊖ g) ⁢ (k) = min ⁢ { f ⁡ (k + s) + g ⁡ (s) ❘ (k + s) ∈ D f, s ∈ D g }, where f represents a signal of the incoming information states, g is a structuring element that lies in domains D f, D g, s is a sampling constant of the signal, and k is a sampling constant of the structuring element, the calculating of the first resolution MMG comprising subtracting a dilated signal from an eroded signal, the subtracting of the dilated signal from the eroded signal comprising using the following equation: ∇ i w =(f⊕g) w −(f⊖g) w, where w represents a resolution level, the second threshold being greater than the first threshold, the first threshold being based on a topology of a communication graph between the agents of the power system, the second threshold being based on the topology of the communication graph between the agents of the power system, the power system comprising at least one of a nanogrid, a microgrid, and a power system area, and the agents of the power system comprising at least 20 agents."
  ],
  "description_excerpt": "Power system control and management architectures are typically built based on centralized algorithms. The future growth of distributed energy resources (DERs) deployment encourages the power system to depend mainly on distributed control/management algorithms. A main drawback of distributed algorithms is their dependence on limited peer-to-peer information broadcast. Each controller or manager is implemented locally to satisfy certain local objectives and is limited with respect to any global objective, which is transmitted by only the information from its neighbors. While the local objective can be satisfying local energy balance, stabilizing local voltage, or maximizing local profits, the global objective may be equal power sharing among different DERs, voltage stabilization at the point of common coupling (PCC), or synchronizing distributed energy entities.\n\nEmbodiments of the subject invention provide novel and advantageous systems and methods for security in power systems. A security-aware distributed control framework for resilient operation of power systems can detect and mitigate different types of attacks that might target power systems. The framework is fast, reliable, and scalable and is able to capture cyber system dynamical features and discriminate between a normal change in control law (cyber system behavior) and a malicious control agent (attacker behavior). The framework can discover a change in features of transmitted data from neighbor agents, discard an infected agent, and achieve an updated consensus protocol agreement while satisfying a control system objective.",
  "cpc": [
    "H04L 63/1416",
    "G01R 31/08",
    "G01R 31/086",
    "G06F 18/10",
    "G06F 18/29",
    "G06F 2218/08",
    "G06K 9/6298",
    "H02J 3/17",
    "H02J 3/32",
    "H02J 3/48",
    "H04L 41/0631",
    "H04L 63/20",
    "Y02B 70/3225",
    "Y04S 20/222",
    "Y04S 40/00",
    "Y04S 40/166",
    "Y04S 40/20",
    "Y04S 40/24"
  ],
  "ipc": [
    "G06F 18/10",
    "H04L 12/24",
    "H04L 29/06"
  ],
  "assignees": [
    "Florida International University FIU"
  ],
  "inventors": [
    "Ahmed Aly Saad Ahmed",
    "Samy Gamal Faddel Mohamed",
    "Osama MOHAMMED"
  ],
  "filing_date": "2020-02-05",
  "publication_date": "2020-06-16",
  "grant_date": "2020-06-16",
  "priority_date": "2020-02-05",
  "application_number": "US-202016782638-A",
  "family_id": "71075088",
  "cited_by_count": 2,
  "citations": [
    "US20040193329A1",
    "US20090299542A1",
    "US20110288692A1",
    "US20130221977A1",
    "US20130003238A1",
    "US20130054162A1",
    "US20130193766A1",
    "US20120266209A1",
    "US20140058689A1",
    "US9615269B2",
    "US20160124031A1",
    "US20160366586A1",
    "US20170093889A1",
    "US10564247B2",
    "US20170307676A1",
    "US20180115561A1",
    "US20180262525A1"
  ]
}